Search

Beyond maximum likelihood estimation (MLE), the standard objective of a language model (LM) that optimizes good examples probabilities, many studies have explored ways that also penalize bad examples for enhancing the quality of output distribution, including unlikelihood training, exponential maximizing average treatment effect (ExMATE), and direct preference optimization (DPO). To systematically compare these methods and further provide a unified recipe for LM optimization, in this paper, we present a unique angle of gradient analysis of loss functions that simultaneously reward good examples and penalize bad ones in LMs. Through both mathematical results and experiments on CausalDialogue and Anthropic HH-RLHF datasets, we identify distinct functional characteristics among these methods. We find that ExMATE serves as a superior surrogate for MLE, and that combining DPO with ExMATE instead of MLE further enhances both the statistical (5-7%) and generative (+18% win rate) performance.

Knowledge editing has been increasingly adopted to correct the false or outdated knowledge in Large Language Models (LLMs). Meanwhile, one critical but under-explored question is: can knowledge editing be used to inject harm into LLMs? In this paper, we propose to reformulate knowledge editing as a new type of safety threat for LLMs, namely Editing Attack, and conduct a systematic investigation with a newly constructed dataset EditAttack. Specifically, we focus on two typical safety risks of Editing Attack including Misinformation Injection and Bias Injection. For the risk of misinformation injection, we first categorize it into commonsense misinformation injection and long-tail misinformation injection. Then, we find that editing attacks can inject both types of misinformation into LLMs, and the effectiveness is particularly high for commonsense misinformation injection. For the risk of bias injection, we discover that not only can biased sentences be injected into LLMs with high effectiveness, but also one single biased sentence injection can cause a bias increase in general outputs of LLMs, which are even highly irrelevant to the injected sentence, indicating a catastrophic impact on the overall fairness of LLMs. Then, we further illustrate the high stealthiness of editing attacks, measured by their impact on the general knowledge and reasoning capacities of LLMs, and show the hardness of defending editing attacks with empirical evidence. Our discoveries demonstrate the emerging misuse risks of knowledge editing techniques on compromising the safety alignment of LLMs and the feasibility of disseminating misinformation or bias with LLMs as new channels., Comment: The first two authors contributed equally. 9 pages for main paper, 36 pages including appendix. The code, results, dataset for this paper and more resources are on the project website: https://llm-editing.github.io

Modern language models (LMs) pose a new challenge in capability assessment. Static benchmarks inevitably saturate without providing confidence in the deployment tolerances of LM-based systems, but developers nonetheless claim that their models have generalized traits such as reasoning or open-domain language understanding based on these flawed metrics. The science and practice of LMs requires a new approach to benchmarking which measures specific capabilities with dynamic assessments. To be confident in our metrics, we need a new discipline of model metrology -- one which focuses on how to generate benchmarks that predict performance under deployment. Motivated by our evaluation criteria, we outline how building a community of model metrology practitioners -- one focused on building tools and studying how to measure system capabilities -- is the best way to meet these needs to and add clarity to the AI discussion., Comment: Conference paper at COLM 2024

Despite the proven utility of large language models (LLMs) in real-world applications, there remains a lack of understanding regarding how they leverage their large-scale pretraining text corpora to achieve such capabilities. In this work, we investigate the interplay between generalization and memorization in pretrained LLMs at scale, through a comprehensive $n$-gram analysis of their training data. Our experiments focus on three general task types: translation, question-answering, and multiple-choice reasoning. With various sizes of open-source LLMs and their pretraining corpora, we observe that as the model size increases, the task-relevant $n$-gram pair data becomes increasingly important, leading to improved task performance, decreased memorization, stronger generalization, and emergent abilities. Our results support the hypothesis that LLMs' capabilities emerge from a delicate balance of memorization and generalization with sufficient task-related pretraining data, and point the way to larger-scale analyses that could further improve our understanding of these models., Comment: ICML FM-Wild workshop version

Question answering based on retrieval augmented generation (RAG-QA) is an important research topic in NLP and has a wide range of real-world applications. However, most existing datasets for this task are either constructed using a single source corpus or consist of short extractive answers, which fall short of evaluating large language model (LLM) based RAG-QA systems on cross-domain generalization. To address these limitations, we create Long-form RobustQA (LFRQA), a new dataset comprising human-written long-form answers that integrate short extractive answers from multiple documents into a single, coherent narrative, covering 26K queries and large corpora across seven different domains. We further propose RAG-QA Arena by directly comparing model-generated answers against LFRQA's answers using LLMs as evaluators. We show via extensive experiments that RAG-QA Arena and human judgments on answer quality are highly correlated. Moreover, only 41.3% of the most competitive LLM's answers are preferred to LFRQA's answers, demonstrating RAG-QA Arena as a challenging evaluation platform for future research.

To enhance Large Language Model (LLM) capabilities, multi-agent debates have been introduced, where multiple LLMs discuss solutions to a problem over several rounds of debate. However, LLMs often produce incorrect responses that appear deceptively confident, which can mislead other agents. This is partly because agents do not express their confidence levels during standard debates. To address this, we introduce DebUnc, a multi-agent debate framework that uses uncertainty metrics to assess agent confidence levels. We adapted the LLM attention mechanism to adjust token weights based on confidence levels and also explored using textual prompts to convey confidence. Our evaluations across various benchmarks show that attention-based methods are particularly effective, and that as uncertainty metrics evolve, performance will continue to increase. The code is available at https://github.com/lukeyoffe/debunc

The rapid advancement of Large Language Models (LLMs) and Large Multimodal Models (LMMs) has heightened the demand for AI-based scientific assistants capable of understanding scientific articles and figures. Despite progress, there remains a significant gap in evaluating models' comprehension of professional, graduate-level, and even PhD-level scientific content. Current datasets and benchmarks primarily focus on relatively simple scientific tasks and figures, lacking comprehensive assessments across diverse advanced scientific disciplines. To bridge this gap, we collected a multimodal, multidisciplinary dataset from open-access scientific articles published in Nature Communications journals. This dataset spans 72 scientific disciplines, ensuring both diversity and quality. We created benchmarks with various tasks and settings to comprehensively evaluate LMMs' capabilities in understanding scientific figures and content. Our evaluation revealed that these tasks are highly challenging: many open-source models struggled significantly, and even GPT-4V and GPT-4o faced difficulties. We also explored using our dataset as training resources by constructing visual instruction-following data, enabling the 7B LLaVA model to achieve performance comparable to GPT-4V/o on our benchmark. Additionally, we investigated the use of our interleaved article texts and figure images for pre-training LMMs, resulting in improvements on the material generation task. The source dataset, including articles, figures, constructed benchmarks, and visual instruction-following data, is open-sourced., Comment: Code and data are available at https://github.com/Leezekun/MMSci

Vision language models (VLMs) are an exciting emerging class of language models (LMs) that have merged classic LM capabilities with those of image processing systems. However, the ways that these capabilities combine are not always intuitive and warrant direct investigation. One understudied capability in VLMs is visual spatial planning -- the ability to comprehend the spatial arrangements of objects and devise action plans to achieve desired outcomes in visual scenes. In our study, we introduce VSP, a benchmark that 1) evaluates the spatial planning capability in these models in general, and 2) breaks down the visual planning task into finer-grained sub-tasks, including perception and reasoning, and measure the LMs capabilities in these sub-tasks. Our evaluation shows that both open-source and private VLMs fail to generate effective plans for even simple spatial planning tasks. Evaluations on the fine-grained analytical tasks further reveal fundamental deficiencies in the models' visual perception and bottlenecks in reasoning abilities, explaining their worse performance in the general spatial planning tasks. Our work illuminates future directions for improving VLMs' abilities in spatial planning. Our benchmark is publicly available at https://github.com/UCSB-NLP-Chang/Visual-Spatial-Planning.

We present LoCoVQA, a dynamic benchmark generator for evaluating long-context extractive reasoning in vision language models (VLMs). LoCoVQA augments test examples for mathematical reasoning, VQA, and character recognition tasks with increasingly long visual contexts composed of both in-distribution and out-of-distribution distractor images. Across these tasks, a diverse set of VLMs rapidly lose performance as the visual context length grows, often exhibiting a striking logarithmic decay trend. This test assesses how well VLMs can ignore irrelevant information when answering queries -- a task that is quite easy for language models (LMs) in the text domain -- demonstrating that current state-of-the-art VLMs lack this essential capability for many long-context applications., Comment: Under review. Minor errata correction in revision

Large language models (LLMs) have shown remarkable performance on code generation tasks. A recent application of LLMs for code generation is iterative code repair, where a model fixes an incorrect program by rationalizing about errors and generating a new program. However, code repair is primarily studied on high-resource languages like Python, and the framework's efficacy is under-explored on low-resource languages. To apply code repair for low-resource languages, we propose Distilling Low-Resource Repairs (DistiLRR), an approach that transfers the reasoning and code generation ability from a teacher model to a student model. Our results show that DistiLRR consistently outperforms baselines on low-resource languages, but has similar performance on high-resource languages. To investigate this behavior, we perform a further analysis and find that the correlation between rationale quality and code correctness is weaker than previously perceived. We hypothesize this weakness is magnified in low-resource settings where base models lack deep knowledge of a programming language, leading to wavering benefits of code repair between high-resource and low-resource languages.

Counterfactual explanations of Graph Neural Networks (GNNs) offer a powerful way to understand data that can naturally be represented by a graph structure. Furthermore, in many domains, it is highly desirable to derive data-driven global explanations or rules that can better explain the high-level properties of the models and data in question. However, evaluating global counterfactual explanations is hard in real-world datasets due to a lack of human-annotated ground truth, which limits their use in areas like molecular sciences. Additionally, the increasing scale of these datasets provides a challenge for random search-based methods. In this paper, we develop a novel global explanation model RLHEX for molecular property prediction. It aligns the counterfactual explanations with human-defined principles, making the explanations more interpretable and easy for experts to evaluate. RLHEX includes a VAE-based graph generator to generate global explanations and an adapter to adjust the latent representation space to human-defined principles. Optimized by Proximal Policy Optimization (PPO), the global explanations produced by RLHEX cover 4.12% more input graphs and reduce the distance between the counterfactual explanation set and the input set by 0.47% on average across three molecular datasets. RLHEX provides a flexible framework to incorporate different human-designed principles into the counterfactual explanation generation process, aligning these explanations with domain expertise. The code and data are released at https://github.com/dqwang122/RLHEX., Comment: Accepted by KDD 2024

Direct alignment from preferences (DAP) has emerged as a promising paradigm for aligning large language models (LLMs) to human desiderata from pre-collected, offline preference datasets. While recent studies indicate that existing offline DAP methods can directly benefit from online training samples, we highlight the need to develop specific online DAP algorithms to fully harness the power of online training. Specifically, we identify that the learned LLM should adhere to the proximity of the behavior LLM, which collects the training samples. To this end, we propose online Preference Optimization in proximity to the Behavior LLM (BPO), emphasizing the importance of constructing a proper trust region for LLM alignment. We conduct extensive experiments to validate the effectiveness and applicability of our approach by integrating it with various DAP methods, resulting in significant performance improvements across a wide range of tasks when training with the same amount of preference data. Even when only introducing one additional data collection phase, our online BPO improves its offline DAP baseline from 72.0% to 80.2% on TL;DR and from 82.2% to 89.1% on Anthropic Helpfulness in terms of win rate against human reference text., Comment: Wenda Xu and Jiachen Li contributed equally

Recent breakthroughs in vision-language models (VLMs) emphasize the necessity of benchmarking human preferences in real-world multimodal interactions. To address this gap, we launched WildVision-Arena (WV-Arena), an online platform that collects human preferences to evaluate VLMs. We curated WV-Bench by selecting 500 high-quality samples from 8,000 user submissions in WV-Arena. WV-Bench uses GPT-4 as the judge to compare each VLM with Claude-3-Sonnet, achieving a Spearman correlation of 0.94 with the WV-Arena Elo. This significantly outperforms other benchmarks like MMVet, MMMU, and MMStar. Our comprehensive analysis of 20K real-world interactions reveals important insights into the failure cases of top-performing VLMs. For example, we find that although GPT-4V surpasses many other models like Reka-Flash, Opus, and Yi-VL-Plus in simple visual recognition and reasoning tasks, it still faces challenges with subtle contextual cues, spatial reasoning, visual imagination, and expert domain knowledge. Additionally, current VLMs exhibit issues with hallucinations and safety when intentionally provoked. We are releasing our chat and feedback data to further advance research in the field of VLMs., Comment: link: https://hf.co/spaces/WildVision/vision-arena

Video generation has many unique challenges beyond those of image generation. The temporal dimension introduces extensive possible variations across frames, over which consistency and continuity may be violated. In this study, we move beyond evaluating simple actions and argue that generated videos should incorporate the emergence of new concepts and their relation transitions like in real-world videos as time progresses. To assess the Temporal Compositionality of video generation models, we propose TC-Bench, a benchmark of meticulously crafted text prompts, corresponding ground truth videos, and robust evaluation metrics. The prompts articulate the initial and final states of scenes, effectively reducing ambiguities for frame development and simplifying the assessment of transition completion. In addition, by collecting aligned real-world videos corresponding to the prompts, we expand TC-Bench's applicability from text-conditional models to image-conditional ones that can perform generative frame interpolation. We also develop new metrics to measure the completeness of component transitions in generated videos, which demonstrate significantly higher correlations with human judgments than existing metrics. Our comprehensive experimental results reveal that most video generators achieve less than 20% of the compositional changes, highlighting enormous space for future improvement. Our analysis indicates that current video generation models struggle to interpret descriptions of compositional changes and synthesize various components across different time steps.

Multimodal Language Language Models (MLLMs) demonstrate the emerging abilities of "world models" -- interpreting and reasoning about complex real-world dynamics. To assess these abilities, we posit videos are the ideal medium, as they encapsulate rich representations of real-world dynamics and causalities. To this end, we introduce MMWorld, a new benchmark for multi-discipline, multi-faceted multimodal video understanding. MMWorld distinguishes itself from previous video understanding benchmarks with two unique advantages: (1) multi-discipline, covering various disciplines that often require domain expertise for comprehensive understanding; (2) multi-faceted reasoning, including explanation, counterfactual thinking, future prediction, etc. MMWorld consists of a human-annotated dataset to evaluate MLLMs with questions about the whole videos and a synthetic dataset to analyze MLLMs within a single modality of perception. Together, MMWorld encompasses 1,910 videos across seven broad disciplines and 69 subdisciplines, complete with 6,627 question-answer pairs and associated captions. The evaluation includes 2 proprietary and 10 open-source MLLMs, which struggle on MMWorld (e.g., GPT-4V performs the best with only 52.3\% accuracy), showing large room for improvement. Further ablation studies reveal other interesting findings such as models' different skill sets from humans. We hope MMWorld can serve as an essential step towards world model evaluation in videos.

We present a novel task and benchmark for evaluating the ability of text-to-image(T2I) generation models to produce images that align with commonsense in real life, which we call Commonsense-T2I. Given two adversarial text prompts containing an identical set of action words with minor differences, such as "a lightbulb without electricity" v.s. "a lightbulb with electricity", we evaluate whether T2I models can conduct visual-commonsense reasoning, e.g. produce images that fit "the lightbulb is unlit" vs. "the lightbulb is lit" correspondingly. Commonsense-T2I presents an adversarial challenge, providing pairwise text prompts along with expected outputs. The dataset is carefully hand-curated by experts and annotated with fine-grained labels, such as commonsense type and likelihood of the expected outputs, to assist analyzing model behavior. We benchmark a variety of state-of-the-art (sota) T2I models and surprisingly find that, there is still a large gap between image synthesis and real life photos--even the DALL-E 3 model could only achieve 48.92% on Commonsense-T2I, and the stable diffusion XL model only achieves 24.92% accuracy. Our experiments show that GPT-enriched prompts cannot solve this challenge, and we include a detailed analysis about possible reasons for such deficiency. We aim for Commonsense-T2I to serve as a high-quality evaluation benchmark for T2I commonsense checking, fostering advancements in real life image generation., Comment: COLM 2024, Project Url: https://zeyofu.github.io/CommonsenseT2I/

Instruction Fine-Tuning (IFT) significantly enhances the zero-shot capabilities of pretrained Large Language Models (LLMs). While coding data is known to boost reasoning abilities during LLM pretraining, its role in activating internal reasoning capacities during IFT remains understudied. This paper investigates a key question: How does coding data impact LLMs' reasoning capacities during the IFT stage? To explore this, we thoroughly examine the impact of coding data across different coding data proportions, model families, sizes, and reasoning domains, from various perspectives. Specifically, we create three IFT datasets with increasing coding data proportions, fine-tune six LLM backbones across different families and scales on these datasets, evaluate the tuned models' performance across twelve tasks in three reasoning domains, and analyze the outcomes from three broad-to-granular perspectives: overall, domain-level, and task-specific. Our holistic analysis provides valuable insights in each perspective. First, coding data tuning enhances the overall reasoning capabilities of LLMs across different model families and scales. Moreover, the effect of coding data varies among different domains but shows consistent trends across model families and scales within each domain. Additionally, coding data generally yields comparable task-specific benefits across different model families, with the optimal coding data proportions in IFT datasets being task-specific.

Diffusion-based text-to-video (T2V) models have achieved significant success but continue to be hampered by the slow sampling speed of their iterative sampling processes. To address the challenge, consistency models have been proposed to facilitate fast inference, albeit at the cost of sample quality. In this work, we aim to break the quality bottleneck of a video consistency model (VCM) to achieve $\textbf{both fast and high-quality video generation}$. We introduce T2V-Turbo, which integrates feedback from a mixture of differentiable reward models into the consistency distillation (CD) process of a pre-trained T2V model. Notably, we directly optimize rewards associated with single-step generations that arise naturally from computing the CD loss, effectively bypassing the memory constraints imposed by backpropagating gradients through an iterative sampling process. Remarkably, the 4-step generations from our T2V-Turbo achieve the highest total score on VBench, even surpassing Gen-2 and Pika. We further conduct human evaluations to corroborate the results, validating that the 4-step generations from our T2V-Turbo are preferred over the 50-step DDIM samples from their teacher models, representing more than a tenfold acceleration while improving video generation quality., Comment: Project page: https://t2v-turbo.github.io/

Topic models have been evolving rapidly over the years, from conventional to recent neural models. However, existing topic models generally struggle with either effectiveness, efficiency, or stability, highly impeding their practical applications. In this paper, we propose FASTopic, a fast, adaptive, stable, and transferable topic model. FASTopic follows a new paradigm: Dual Semantic-relation Reconstruction (DSR). Instead of previous conventional, neural VAE-based or clustering-based methods, DSR discovers latent topics by reconstruction through modeling the semantic relations among document, topic, and word embeddings. This brings about a neat and efficient topic modeling framework. We further propose a novel Embedding Transport Plan (ETP) method. Rather than early straightforward approaches, ETP explicitly regularizes the semantic relations as optimal transport plans. This addresses the relation bias issue and thus leads to effective topic modeling. Extensive experiments on benchmark datasets demonstrate that our FASTopic shows superior effectiveness, efficiency, adaptivity, stability, and transferability, compared to state-of-the-art baselines across various scenarios. Our code is available at https://github.com/bobxwu/FASTopic .

The rapid progress in Multimodal Large Language Models (MLLMs) has significantly advanced their ability to process and understand complex visual and textual information. However, the integration of multiple images and extensive textual contexts remains a challenge due to the inherent limitation of the models' capacity to handle long input sequences efficiently. In this paper, we introduce SEEKER, a multimodal large language model designed to tackle this issue. SEEKER aims to optimize the compact encoding of long text by compressing the text sequence into the visual pixel space via images, enabling the model to handle long text within a fixed token-length budget efficiently. Our empirical experiments on six long-context multimodal tasks demonstrate that SEEKER can leverage fewer image tokens to convey the same amount of textual information compared with the OCR-based approach, and is more efficient in understanding long-form multimodal input and generating long-form textual output, outperforming all existing proprietary and open-source MLLMs by large margins.

In the fast-evolving domain of artificial intelligence, large language models (LLMs) such as GPT-3 and GPT-4 are revolutionizing the landscapes of finance, healthcare, and law: domains characterized by their reliance on professional expertise, challenging data acquisition, high-stakes, and stringent regulatory compliance. This survey offers a detailed exploration of the methodologies, applications, challenges, and forward-looking opportunities of LLMs within these high-stakes sectors. We highlight the instrumental role of LLMs in enhancing diagnostic and treatment methodologies in healthcare, innovating financial analytics, and refining legal interpretation and compliance strategies. Moreover, we critically examine the ethics for LLM applications in these fields, pointing out the existing ethical concerns and the need for transparent, fair, and robust AI systems that respect regulatory norms. By presenting a thorough review of current literature and practical applications, we showcase the transformative impact of LLMs, and outline the imperative for interdisciplinary cooperation, methodological advancements, and ethical vigilance. Through this lens, we aim to spark dialogue and inspire future research dedicated to maximizing the benefits of LLMs while mitigating their risks in these precision-dependent sectors. To facilitate future research on LLMs in these critical societal domains, we also initiate a reading list that tracks the latest advancements under this topic, which will be continually updated: \url{https://github.com/czyssrs/LLM_X_papers}., Comment: 35 pages, 6 figures

Recent advancements in instruction-following models have made user interactions with models more user-friendly and efficient, broadening their applicability. In graphic design, non-professional users often struggle to create visually appealing layouts due to limited skills and resources. In this work, we introduce a novel multimodal instruction-following framework for layout planning, allowing users to easily arrange visual elements into tailored layouts by specifying canvas size and design purpose, such as for book covers, posters, brochures, or menus. We developed three layout reasoning tasks to train the model in understanding and executing layout instructions. Experiments on two benchmarks show that our method not only simplifies the design process for non-professionals but also surpasses the performance of few-shot GPT-4V models, with mIoU higher by 12% on Crello. This progress highlights the potential of multimodal instruction-following models to automate and simplify the design process, providing an approachable solution for a wide range of design tasks on visually-rich documents.

While Ferret seamlessly integrates regional understanding into the Large Language Model (LLM) to facilitate its referring and grounding capability, it poses certain limitations: constrained by the pre-trained fixed visual encoder and failed to perform well on broader tasks. In this work, we unveil Ferret-v2, a significant upgrade to Ferret, with three key designs. (1) Any resolution grounding and referring: A flexible approach that effortlessly handles higher image resolution, improving the model's ability to process and understand images in greater detail. (2) Multi-granularity visual encoding: By integrating the additional DINOv2 encoder, the model learns better and diverse underlying contexts for global and fine-grained visual information. (3) A three-stage training paradigm: Besides image-caption alignment, an additional stage is proposed for high-resolution dense alignment before the final instruction tuning. Experiments show that Ferret-v2 provides substantial improvements over Ferret and other state-of-the-art methods, thanks to its high-resolution scaling and fine-grained visual processing., Comment: Preprint. 14 pages, 4 figures

With advances in the quality of text-to-image (T2I) models has come interest in benchmarking their prompt faithfulness-the semantic coherence of generated images to the prompts they were conditioned on. A variety of T2I faithfulness metrics have been proposed, leveraging advances in cross-modal embeddings and vision-language models (VLMs). However, these metrics are not rigorously compared and benchmarked, instead presented with correlation to human Likert scores over a set of easy-to-discriminate images against seemingly weak baselines. We introduce T2IScoreScore (TS2), a curated set of semantic error graphs containing a prompt and a set increasingly erroneous images. These allow us to rigorously judge whether a given prompt faithfulness metric can correctly order images with respect to their objective error count and significantly discriminate between different error nodes, using meta-metric scores derived from established statistical tests. Surprisingly, we find that the state-of-the-art VLM-based metrics (e.g., TIFA, DSG, LLMScore, VIEScore) we tested fail to significantly outperform simple (and supposedly worse) feature-based metrics like CLIPScore, particularly on a hard subset of naturally-occurring T2I model errors. TS2 will enable the development of better T2I prompt faithfulness metrics through more rigorous comparison of their conformity to expected orderings and separations under objective criteria., Comment: 10 pages main, 12 pages appendices, 13 figures, 3 tables

As large language models (LLMs) become easily accessible nowadays, the trade-off between safety and helpfulness can significantly impact user experience. A model that prioritizes safety will cause users to feel less engaged and assisted while prioritizing helpfulness will potentially cause harm. Possible harms include teaching people how to build a bomb, exposing youth to inappropriate content, and hurting users' mental health. In this work, we propose to balance safety and helpfulness in diverse use cases by controlling both attributes in LLM. We explore training-free and fine-tuning methods that do not require extra human annotations and analyze the challenges of controlling safety and helpfulness in LLMs. Our experiments demonstrate that our method can rewind a learned model and unlock its controllability.

Benchmarks of the multilingual capabilities of text-to-image (T2I) models compare generated images prompted in a test language to an expected image distribution over a concept set. One such benchmark, "Conceptual Coverage Across Languages" (CoCo-CroLa), assesses the tangible noun inventory of T2I models by prompting them to generate pictures from a concept list translated to seven languages and comparing the output image populations. Unfortunately, we find that this benchmark contains translation errors of varying severity in Spanish, Japanese, and Chinese. We provide corrections for these errors and analyze how impactful they are on the utility and validity of CoCo-CroLa as a benchmark. We reassess multiple baseline T2I models with the revisions, compare the outputs elicited under the new translations to those conditioned on the old, and show that a correction's impactfulness on the image-domain benchmark results can be predicted in the text domain with similarity scores. Our findings will guide the future development of T2I multilinguality metrics by providing analytical tools for practical translation decisions., Comment: NAACL 2024 Main Conference

Latent Consistency Distillation (LCD) has emerged as a promising paradigm for efficient text-to-image synthesis. By distilling a latent consistency model (LCM) from a pre-trained teacher latent diffusion model (LDM), LCD facilitates the generation of high-fidelity images within merely 2 to 4 inference steps. However, the LCM's efficient inference is obtained at the cost of the sample quality. In this paper, we propose compensating the quality loss by aligning LCM's output with human preference during training. Specifically, we introduce Reward Guided LCD (RG-LCD), which integrates feedback from a reward model (RM) into the LCD process by augmenting the original LCD loss with the objective of maximizing the reward associated with LCM's single-step generation. As validated through human evaluation, when trained with the feedback of a good RM, the 2-step generations from our RG-LCM are favored by humans over the 50-step DDIM samples from the teacher LDM, representing a 25 times inference acceleration without quality loss. As directly optimizing towards differentiable RMs can suffer from over-optimization, we overcome this difficulty by proposing the use of a latent proxy RM (LRM). This novel component serves as an intermediary, connecting our LCM with the RM. Empirically, we demonstrate that incorporating the LRM into our RG-LCD successfully avoids high-frequency noise in the generated images, contributing to both improved FID on MS-COCO and a higher HPSv2.1 score on HPSv2's test set, surpassing those achieved by the baseline LCM., Comment: Project page: https://rg-lcd.github.io/

Knowledge editing aims to inject knowledge updates into language models to keep them correct and up-to-date. However, its current evaluation strategies are notably impractical: they solely update with well-curated structured facts (triplets with subjects, relations, and objects), whereas real-world knowledge updates commonly emerge in unstructured texts like news articles. In this paper, we propose a new benchmark, Unstructured Knowledge Editing (UKE). It evaluates editing performance directly using unstructured texts as knowledge updates, termed unstructured facts. Hence UKE avoids the laborious construction of structured facts and enables efficient and responsive knowledge editing, becoming a more practical benchmark. We conduct extensive experiments on newly built datasets and demonstrate that UKE poses a significant challenge to state-of-the-art knowledge editing methods, resulting in their critical performance declines. We further show that this challenge persists even if we extract triplets as structured facts. Our analysis discloses key insights to motivate future research in UKE for more practical knowledge editing.

How can large language models (LLMs) process and translate endangered languages? Many languages lack a large corpus to train a decent LLM; therefore existing LLMs rarely perform well in unseen, endangered languages. On the contrary, we observe that 2000 endangered languages, though without a large corpus, have a grammar book or a dictionary. We propose LINGOLLM, a training-free approach to enable an LLM to process unseen languages that hardly occur in its pre-training. Our key insight is to demonstrate linguistic knowledge of an unseen language in an LLM's prompt, including a dictionary, a grammar book, and morphologically analyzed input text. We implement LINGOLLM on top of two models, GPT-4 and Mixtral, and evaluate their performance on 5 tasks across 8 endangered or low-resource languages. Our results show that LINGOLLM elevates translation capability from GPT-4's 0 to 10.5 BLEU for 10 language directions. Our findings demonstrate the tremendous value of linguistic knowledge in the age of LLMs for endangered languages. Our data, code, and model generations can be found at https://github.com/LLiLab/llm4endangeredlang.

A major factor in the recent success of large language models is the use of enormous and ever-growing text datasets for unsupervised pre-training. However, naively training a model on all available data may not be optimal (or feasible), as the quality of available text data can vary. Filtering out data can also decrease the carbon footprint and financial costs of training models by reducing the amount of training required. Data selection methods aim to determine which candidate data points to include in the training dataset and how to appropriately sample from the selected data points. The promise of improved data selection methods has caused the volume of research in the area to rapidly expand. However, because deep learning is mostly driven by empirical evidence and experimentation on large-scale data is expensive, few organizations have the resources for extensive data selection research. Consequently, knowledge of effective data selection practices has become concentrated within a few organizations, many of which do not openly share their findings and methodologies. To narrow this gap in knowledge, we present a comprehensive review of existing literature on data selection methods and related research areas, providing a taxonomy of existing approaches. By describing the current landscape of research, this work aims to accelerate progress in data selection by establishing an entry point for new and established researchers. Additionally, throughout this review we draw attention to noticeable holes in the literature and conclude the paper by proposing promising avenues for future research., Comment: Paper list available at https://github.com/alon-albalak/data-selection-survey

Recent studies show that large language models (LLMs) improve their performance through self-feedback on certain tasks while degrade on others. We discovered that such a contrary is due to LLM's bias in evaluating their own output. In this paper, we formally define LLM's self-bias - the tendency to favor its own generation - using two statistics. We analyze six LLMs (GPT-4, GPT-3.5, Gemini, LLaMA2, Mixtral and DeepSeek) on translation, constrained text generation, and mathematical reasoning tasks. We find that self-bias is prevalent in all examined LLMs across multiple languages and tasks. Our analysis reveals that while the self-refine pipeline improves the fluency and understandability of model outputs, it further amplifies self-bias. To mitigate such biases, we discover that larger model size and external feedback with accurate assessment can significantly reduce bias in the self-refine pipeline, leading to actual performance improvement in downstream tasks. The code and data are released at https://github.com/xu1998hz/llm_self_bias.

Pre-trained language models (LMs) are able to perform complex reasoning without explicit fine-tuning. To understand how pre-training with a next-token prediction objective contributes to the emergence of such reasoning capability, we propose that we can view an LM as deriving new conclusions by aggregating indirect reasoning paths seen at pre-training time. We found this perspective effective in two important cases of reasoning: logic reasoning with knowledge graphs (KGs) and chain-of-thought (CoT) reasoning. More specifically, we formalize the reasoning paths as random walk paths on the knowledge/reasoning graphs. Analyses of learned LM distributions suggest that a weighted sum of relevant random walk path probabilities is a reasonable way to explain how LMs reason. Experiments and analysis on multiple KG and CoT datasets reveal the effect of training on random walk paths and suggest that augmenting unlabeled random walk reasoning paths can improve real-world multi-step reasoning performance. code: https://github.com/WANGXinyiLinda/LM_random_walk, Comment: Accepted to ICML 2024

Large language models (LLMs) are vulnerable to jailbreak attacks - resulting in harmful, unethical, or biased text generations. However, existing jailbreaking methods are computationally costly. In this paper, we propose the weak-to-strong jailbreaking attack, an efficient method to attack aligned LLMs to produce harmful text. Our key intuition is based on the observation that jailbroken and aligned models only differ in their initial decoding distributions. The weak-to-strong attack's key technical insight is using two smaller models (a safe and an unsafe one) to adversarially modify a significantly larger safe model's decoding probabilities. We evaluate the weak-to-strong attack on 5 diverse LLMs from 3 organizations. The results show our method can increase the misalignment rate to over 99% on two datasets with just one forward pass per example. Our study exposes an urgent safety issue that needs to be addressed when aligning LLMs. As an initial attempt, we propose a defense strategy to protect against such attacks, but creating more advanced defenses remains challenging. The code for replicating the method is available at https://github.com/XuandongZhao/weak-to-strong

As the number of accepted papers at AI and ML conferences reaches into the thousands, it has become unclear how researchers access and read research publications. In this paper, we investigate the role of social media influencers in enhancing the visibility of machine learning research, particularly the citation counts of papers they share. We have compiled a comprehensive dataset of over 8,000 papers, spanning tweets from December 2018 to October 2023, alongside controls precisely matched by 9 key covariates. Our statistical and causal inference analysis reveals a significant increase in citations for papers endorsed by these influencers, with median citation counts 2-3 times higher than those of the control group. Additionally, the study delves into the geographic, gender, and institutional diversity of highlighted authors. Given these findings, we advocate for a responsible approach to curation, encouraging influencers to uphold the journalistic standard that includes showcasing diverse research topics, authors, and institutions., Comment: 15 Pages, 22 Figures, ICML 2024

The data used to pretrain large language models has a decisive impact on a model's downstream performance, which has led to a large body of work on data selection methods that aim to automatically determine the most suitable data to use for pretraining. Existing data selection methods suffer from slow and computationally expensive processes, a problem amplified by the increasing size of models and of pretraining datasets. Data mixing, on the other hand, reduces the complexity of data selection by grouping data points together and determining sampling probabilities across entire groups. However, data mixing proportions are typically fixed before training and therefore cannot adapt to changing training dynamics. To address these limitations, we develop an efficient algorithm for Online Data Mixing (ODM) that combines elements from both data selection and data mixing. Based on multi-armed bandit algorithms, our online approach optimizes the data mixing proportions during training. Remarkably, our method trains a model that reaches the final perplexity of the next best method with 19\% fewer training iterations, and improves performance on the 5-shot MMLU benchmark by 1.9% relative accuracy, while adding negligible wall-clock time during pretraining.

Recent multimodal large language models (MLLMs) have shown promising instruction following capabilities on vision-language tasks. In this work, we introduce VISUAL MODALITY INSTRUCTION (VIM), and investigate how well multimodal models can understand textual instructions provided in pixels, despite not being explicitly trained on such data during pretraining or fine-tuning. We adapt VIM to eight benchmarks, including OKVQA, MM-Vet, MathVista, MMMU, and probe diverse MLLMs in both the text-modality instruction (TEM) setting and VIM setting. Notably, we observe a significant performance disparity between the original TEM and VIM settings for open-source MLLMs, indicating that open-source MLLMs face greater challenges when text instruction is presented solely in image form. To address this issue, we train v-MLLM, a generalizable model that is capable to conduct robust instruction following in both text-modality and visual-modality instructions., Comment: Github: https://github.com/VIM-Bench/VIM_TOOL, Model and Data: https://huggingface.co/VIM-Bench

Recent large language models (LLM) are leveraging human feedback to improve their generation quality. However, human feedback is costly to obtain, especially during inference. In this work, we propose LLMRefine, an inference time optimization method to refine LLM's output. The core idea is to use a learned fine-grained feedback model to pinpoint defects and guide LLM to refine them iteratively. Using original LLM as a proposal of edits, LLMRefine searches for defect-less text via simulated annealing, trading off the exploration and exploitation. We conduct experiments on three text generation tasks, including machine translation, long-form question answering (QA), and topical summarization. LLMRefine consistently outperforms all baseline approaches, achieving improvements up to 1.7 MetricX points on translation tasks, 8.1 ROUGE-L on ASQA, 2.2 ROUGE-L on topical summarization., Comment: Accepted to NAACL 2024

Automatically evaluating vision-language tasks is challenging, especially when it comes to reflecting human judgments due to limitations in accounting for fine-grained details. Although GPT-4V has shown promising results in various multi-modal tasks, leveraging GPT-4V as a generalist evaluator for these tasks has not yet been systematically explored. We comprehensively validate GPT-4V's capabilities for evaluation purposes, addressing tasks ranging from foundational image-to-text and text-to-image synthesis to high-level image-to-image translations and multi-images to text alignment. We employ two evaluation methods, single-answer grading and pairwise comparison, using GPT-4V. Notably, GPT-4V shows promising agreement with humans across various tasks and evaluation methods, demonstrating immense potential for multi-modal LLMs as evaluators. Despite limitations like restricted visual clarity grading and real-world complex reasoning, its ability to provide human-aligned scores enriched with detailed explanations is promising for universal automatic evaluator.

The burgeoning capabilities of advanced large language models (LLMs) such as ChatGPT have led to an increase in synthetic content generation with implications across a variety of sectors, including media, cybersecurity, public discourse, and education. As such, the ability to detect LLMs-generated content has become of paramount importance. We aim to provide a detailed overview of existing detection strategies and benchmarks, scrutinizing their differences and identifying key challenges and prospects in the field, advocating for more adaptable and robust models to enhance detection accuracy. We also posit the necessity for a multi-faceted approach to defend against various attacks to counter the rapidly advancing capabilities of LLMs. To the best of our knowledge, this work is the first comprehensive survey on the detection in the era of LLMs. We hope it will provide a broad understanding of the current landscape of LLMs-generated content detection, offering a guiding reference for researchers and practitioners striving to uphold the integrity of digital information in an era increasingly dominated by synthetic content. The relevant papers are summarized and will be consistently updated at https://github.com/Xianjun-Yang/Awesome_papers_on_LLMs_detection.git., Comment: We will keep updating at https://github.com/Xianjun-Yang/Awesome_papers_on_LLMs_detection.git

Language Models (LMs) have shown impressive performance in various natural language tasks. However, when it comes to natural language reasoning, LMs still face challenges such as hallucination, generating incorrect intermediate reasoning steps, and making mathematical errors. Recent research has focused on enhancing LMs through self-improvement using feedback. Nevertheless, existing approaches relying on a single generic feedback source fail to address the diverse error types found in LM-generated reasoning chains. In this work, we propose Multi-Aspect Feedback, an iterative refinement framework that integrates multiple feedback modules, including frozen LMs and external tools, each focusing on a specific error category. Our experimental results demonstrate the efficacy of our approach to addressing several errors in the LM-generated reasoning chain and thus improving the overall performance of an LM in several reasoning tasks. We see a relative improvement of up to 20% in Mathematical Reasoning and up to 18% in Logical Entailment., Comment: Accepted at EMNLP 2023 Main Conference, Camera Ready

Prompt-based learning has been demonstrated as a compelling paradigm contributing to large language models' tremendous success (LLMs). Inspired by their success in language tasks, existing research has leveraged LLMs in embodied instruction following and task planning. In this work, we tackle the problem of training a robot to understand multimodal prompts, interleaving vision signals with text descriptions. This type of task poses a major challenge to robots' capability to understand the interconnection and complementarity between vision and language signals. In this work, we introduce an effective framework that learns a policy to perform robot manipulation with multimodal prompts from multi-task expert trajectories. Our methods consist of a two-stage training pipeline that performs inverse dynamics pretraining and multi-task finetuning. To facilitate multimodal understanding, we design our multimodal prompt encoder by augmenting a pretrained LM with a residual connection to the visual input and model the dependencies among action dimensions. Empirically, we evaluate the efficacy of our method on the VIMA-BENCH and establish a new state-of-the-art (10% improvement in success rate). Moreover, we demonstrate that our model exhibits remarkable in-context learning ability. Project page: \url{https://midas-icml.github.io/}., Comment: Accepted by ICML 2024. Project page: https://midas-icml.github.io

As large language models are integrated into society, robustness toward a suite of prompts is increasingly important to maintain reliability in a high-variance environment.Robustness evaluations must comprehensively encapsulate the various settings in which a user may invoke an intelligent system. This paper proposes ASSERT, Automated Safety Scenario Red Teaming, consisting of three methods -- semantically aligned augmentation, target bootstrapping, and adversarial knowledge injection. For robust safety evaluation, we apply these methods in the critical domain of AI safety to algorithmically generate a test suite of prompts covering diverse robustness settings -- semantic equivalence, related scenarios, and adversarial. We partition our prompts into four safety domains for a fine-grained analysis of how the domain affects model performance. Despite dedicated safeguards in existing state-of-the-art models, we find statistically significant performance differences of up to 11% in absolute classification accuracy among semantically related scenarios and error rates of up to 19% absolute error in zero-shot adversarial settings, raising concerns for users' physical safety., Comment: In Findings of the 2023 Conference on Empirical Methods in Natural Language Processing

Mental illness remains one of the most critical public health issues of our time, due to the severe scarcity and accessibility limit of professionals. Psychotherapy requires high-level expertise to conduct deep, complex reasoning and analysis on the cognition modeling of the patients. In the era of Large Language Models, we believe it is the right time to develop AI assistance for computational psychotherapy. We study the task of cognitive distortion detection and propose the Diagnosis of Thought (DoT) prompting. DoT performs diagnosis on the patient's speech via three stages: subjectivity assessment to separate the facts and the thoughts; contrastive reasoning to elicit the reasoning processes supporting and contradicting the thoughts; and schema analysis to summarize the cognition schemas. The generated diagnosis rationales through the three stages are essential for assisting the professionals. Experiments demonstrate that DoT obtains significant improvements over ChatGPT for cognitive distortion detection, while generating high-quality rationales approved by human experts., Comment: EMNLP 2023 Findings

Large language models (LLMs) have recently attracted considerable interest for their ability to perform complex reasoning tasks, such as chain-of-thought (CoT) reasoning. However, most of the existing approaches to enhance this ability rely heavily on data-driven methods, while neglecting the structural aspects of the model's reasoning capacity. To encourage a more structural generation of CoT steps, we propose a hierarchical generation scheme: we let the LM generate a planning token at the start of each reasoning step, intuitively serving as a high-level plan of the current step, and add their embeddings to the model parameters. Our approach requires a negligible increase in trainable parameters (0.001%) and can be applied through either full fine-tuning or a more parameter-efficient scheme. We demonstrate our method's effectiveness by applying it to three different LLMs, showing notable accuracy improvements across three math word problem datasets and one multihop QA dataset with respect to standard fine-tuning baselines., Comment: Accepted to COLM 2024

This work proposes a training-free approach for the detection of LLMs-generated codes, mitigating the risks associated with their indiscriminate usage. To the best of our knowledge, our research is the first to investigate zero-shot detection techniques applied to code generated by advanced black-box LLMs like ChatGPT. Firstly, we find that existing training-based or zero-shot text detectors are ineffective in detecting code, likely due to the unique statistical properties found in code structures. We then modify the previous zero-shot text detection method, DetectGPT (Mitchell et al., 2023) by utilizing a surrogate white-box model to estimate the probability of the rightmost tokens, allowing us to identify code snippets generated by language models. Through extensive experiments conducted on the python codes of the CodeContest and APPS dataset, our approach demonstrates its effectiveness by achieving state-of-the-art detection results on text-davinci-003, GPT-3.5, and GPT-4 models. Moreover, our method exhibits robustness against revision attacks and generalizes well to Java codes. We also find that the smaller code language model like PolyCoder-160M performs as a universal code detector, outperforming the billion-scale counterpart. The codes will be available at https://github.com/ Xianjun-Yang/Code_detection.git, Comment: work in progress

Warning: This paper contains examples of harmful language, and reader discretion is recommended. The increasing open release of powerful large language models (LLMs) has facilitated the development of downstream applications by reducing the essential cost of data annotation and computation. To ensure AI safety, extensive safety-alignment measures have been conducted to armor these models against malicious use (primarily hard prompt attack). However, beneath the seemingly resilient facade of the armor, there might lurk a shadow. By simply tuning on 100 malicious examples with 1 GPU hour, these safely aligned LLMs can be easily subverted to generate harmful content. Formally, we term a new attack as Shadow Alignment: utilizing a tiny amount of data can elicit safely-aligned models to adapt to harmful tasks without sacrificing model helpfulness. Remarkably, the subverted models retain their capability to respond appropriately to regular inquiries. Experiments across 8 models released by 5 different organizations (LLaMa-2, Falcon, InternLM, BaiChuan2, Vicuna) demonstrate the effectiveness of shadow alignment attack. Besides, the single-turn English-only attack successfully transfers to multi-turn dialogue and other languages. This study serves as a clarion call for a collective effort to overhaul and fortify the safety of open-source LLMs against malicious attackers., Comment: Work in progress

Instruction-based image editing improves the controllability and flexibility of image manipulation via natural commands without elaborate descriptions or regional masks. However, human instructions are sometimes too brief for current methods to capture and follow. Multimodal large language models (MLLMs) show promising capabilities in cross-modal understanding and visual-aware response generation via LMs. We investigate how MLLMs facilitate edit instructions and present MLLM-Guided Image Editing (MGIE). MGIE learns to derive expressive instructions and provides explicit guidance. The editing model jointly captures this visual imagination and performs manipulation through end-to-end training. We evaluate various aspects of Photoshop-style modification, global photo optimization, and local editing. Extensive experimental results demonstrate that expressive instructions are crucial to instruction-based image editing, and our MGIE can lead to a notable improvement in automatic metrics and human evaluation while maintaining competitive inference efficiency., Comment: ICLR'24 (Spotlight) ; Project at https://mllm-ie.github.io ; Code at https://github.com/tsujuifu/pytorch_mgie

Large language models (LLMs) have demonstrated remarkable performance across a wide array of NLP tasks. However, their efficacy is undermined by undesired and inconsistent behaviors, including hallucination, unfaithful reasoning, and toxic content. A promising approach to rectify these flaws is self-correction, where the LLM itself is prompted or guided to fix problems in its own output. Techniques leveraging automated feedback -- either produced by the LLM itself or some external system -- are of particular interest as they are a promising way to make LLM-based solutions more practical and deployable with minimal human feedback. This paper presents a comprehensive review of this emerging class of techniques. We analyze and taxonomize a wide array of recent work utilizing these strategies, including training-time, generation-time, and post-hoc correction. We also summarize the major applications of this strategy and conclude by discussing future directions and challenges., Comment: Work in Progress. Version 2

Incremental decision making in real-world environments is one of the most challenging tasks in embodied artificial intelligence. One particularly demanding scenario is Vision and Language Navigation~(VLN) which requires visual and natural language understanding as well as spatial and temporal reasoning capabilities. The embodied agent needs to ground its understanding of navigation instructions in observations of a real-world environment like Street View. Despite the impressive results of LLMs in other research areas, it is an ongoing problem of how to best connect them with an interactive visual environment. In this work, we propose VELMA, an embodied LLM agent that uses a verbalization of the trajectory and of visual environment observations as contextual prompt for the next action. Visual information is verbalized by a pipeline that extracts landmarks from the human written navigation instructions and uses CLIP to determine their visibility in the current panorama view. We show that VELMA is able to successfully follow navigation instructions in Street View with only two in-context examples. We further finetune the LLM agent on a few thousand examples and achieve 25%-30% relative improvement in task completion over the previous state-of-the-art for two datasets., Comment: Accepted at AAAI 2024

We propose "Conceptual Coverage Across Languages" (CoCo-CroLa), a technique for benchmarking the degree to which any generative text-to-image system provides multilingual parity to its training language in terms of tangible nouns. For each model we can assess "conceptual coverage" of a given target language relative to a source language by comparing the population of images generated for a series of tangible nouns in the source language to the population of images generated for each noun under translation in the target language. This technique allows us to estimate how well-suited a model is to a target language as well as identify model-specific weaknesses, spurious correlations, and biases without a-priori assumptions. We demonstrate how it can be used to benchmark T2I models in terms of multilinguality, and how despite its simplicity it is a good proxy for impressive generalization., Comment: ACL 2023 main conference; 16 pages, 13 figures