Search

Large language models (LLMs), such as o1 from OpenAI, have demonstrated remarkable reasoning capabilities. o1 generates a long chain-of-thought (LongCoT) before answering a question. LongCoT allows LLMs to analyze problems, devise plans, reflect, and backtrack effectively. These actions empower LLM to solve complex problems. After the release of o1, many teams have attempted to replicate its LongCoT and reasoning capabilities. In terms of methods, they primarily rely on knowledge distillation with data from existing models with LongCoT capacities (e.g., OpenAI-o1, Qwen-QwQ, DeepSeek-R1-Preview), leaving significant uncertainties on systematically developing such reasoning abilities. In terms of data domains, these works focus narrowly on math while a few others include coding, limiting their generalizability. This paper introduces a novel approach to enable LLM's LongCoT capacity without distillation from o1-like models or expensive human annotations, where we bootstrap LongCoT (BOLT) from a standard instruct model. BOLT involves three stages: 1) LongCoT data bootstrapping with in-context learning on a standard instruct model; 2) LongCoT supervised finetuning; 3) online training to further refine LongCoT capacities. In BOLT, only a few in-context examples need to be constructed during the bootstrapping stage; in our experiments, we created 10 examples, demonstrating the feasibility of this approach. We use Llama-3.1-70B-Instruct to bootstrap LongCoT and apply our method to various model scales (7B, 8B, 70B). We achieve impressive performance on a variety of benchmarks, Arena-Hard, MT-Bench, WildBench, ZebraLogic, MATH500, which evaluate diverse task-solving and reasoning capabilities., Comment: 36 pages

We study the geometry of the moduli stack of torsion-free sheaves on ribbons. We introduce a stratification of the stack by the complete type of the sheaves, and we investigate the geometric properties of the strata and their closure relation. Then we describe the irreducible components of the stack, by revealing an interesting trichotomy between Fano, Calabi-Yau and canonically polarized cases. Finally, we describe which strata intersect the (semi)stable locus., Comment: 27 pages

We introduce Reward-Guided Speculative Decoding (RSD), a novel framework aimed at improving the efficiency of inference in large language models (LLMs). RSD synergistically combines a lightweight draft model with a more powerful target model, incorporating a controlled bias to prioritize high-reward outputs, in contrast to existing speculative decoding methods that enforce strict unbiasedness. RSD employs a process reward model to evaluate intermediate decoding steps and dynamically decide whether to invoke the target model, optimizing the trade-off between computational cost and output quality. We theoretically demonstrate that a threshold-based mixture strategy achieves an optimal balance between resource utilization and performance. Extensive evaluations on challenging reasoning benchmarks, including Olympiad-level tasks, show that RSD delivers significant efficiency gains against decoding with the target model only (up to 4.4x fewer FLOPs), while achieving significant better accuracy than parallel decoding method on average (up to +3.5). These results highlight RSD as a robust and cost-effective approach for deploying LLMs in resource-intensive scenarios., Comment: 17 pages

Light propagation through artificially patterned anisotropic materials, such as dielectric metasurfaces, enables precise control of the spatio-vectorial properties of optical fields using highly transparent, thin, and flat optical elements. Liquid-crystal cells are a common realization of such devices. Optical losses are typically assumed to be polarization-independent and are therefore often overlooked in modeling these systems. In this work, we introduce electrically tunable liquid-crystal metasurfaces with patterned birefringence and dichroism, achieved by incorporating dichroic dye molecules into the liquid-crystal mixture. These dye molecules align with the liquid crystal, effectively coupling birefringence and dichroism effects. The behavior of these metasurfaces is described using non-unitary Jones matrices, validated through polarimetric measurements. In the case of devices that are patterned to form polarization gratings, we also characterize the diffraction efficiency as a function of the dichroism and birefringence parameters, which can be tuned jointly by applying an electric field across the cell. This study not only introduces a new class of optical components but also deepens our understanding of light propagation through anisotropic materials, where dichroism can naturally arise from bulk material properties or from reflection and transmission laws at their interfaces.

DarkSide-20k is a novel liquid argon dark matter detector currently under construction at the Laboratori Nazionali del Gran Sasso (LNGS) of the Istituto Nazionale di Fisica Nucleare (INFN) that will push the sensitivity for Weakly Interacting Massive Particle (WIMP) detection into the neutrino fog. The core of the apparatus is a dual-phase Time Projection Chamber (TPC), filled with \SI{50} {tonnes} of low radioactivity underground argon (UAr) acting as the WIMP target. NUV-HD-Cryo Silicon Photomultipliers (SiPM)s designed by Fondazione Bruno Kessler (FBK) (Povo, Trento, Italy) were selected as the photon sensors covering two $10.5~\text{m}^2$ Optical Planes, one at each end of the TPC, and a total of $5~\text{m}^2$ photosensitive surface for the liquid argon veto detectors. This paper describes the Quality Assurance and Quality Control (QA/QC) plan and procedures accompanying the production of FBK~NUV-HD-Cryo SiPM wafers manufactured by LFoundry s.r.l. (Avezzano, AQ, Italy). SiPM characteristics are measured at 77~K at the wafer level with a custom-designed probe station. As of May~2024, 603 of the 1400 production wafers (43\% of the total) for DarkSide-20k were tested, including wafers from all 57 production Lots. The wafer yield is $93.6\pm2.5$\%, which exceeds the 80\% specification defined in the original DarkSide-20k production plan.

Programming based approaches to reasoning tasks have substantially expanded the types of questions models can answer about visual scenes. Yet on benchmark visual reasoning data, when models answer correctly, they produce incorrect programs 33% of the time. These models are often right for the wrong reasons and risk unexpected failures on new data. Unit tests play a foundational role in ensuring code correctness and could be used to repair such failures. We propose Visual Unit Testing (ViUniT), a framework to improve the reliability of visual programs by automatically generating unit tests. In our framework, a unit test is represented as a novel image and answer pair meant to verify the logical correctness of a program produced for a given query. Our method leverages a language model to create unit tests in the form of image descriptions and expected answers and image synthesis to produce corresponding images. We conduct a comprehensive analysis of what constitutes an effective visual unit test suite, exploring unit test generation, sampling strategies, image generation methods, and varying the number of programs and unit tests. Additionally, we introduce four applications of visual unit tests: best program selection, answer refusal, re-prompting, and unsupervised reward formulations for reinforcement learning. Experiments with two models across three datasets in visual question answering and image-text matching demonstrate that ViUniT improves model performance by 11.4%. Notably, it enables 7B open-source models to outperform gpt-4o-mini by an average of 7.7% and reduces the occurrence of programs that are correct for the wrong reasons by 40%.

With the rise of multimodal applications, instruction data has become critical for training multimodal language models capable of understanding complex image-based queries. Existing practices rely on powerful but costly large language models (LLMs) or multimodal language models (MLMs) to produce instruction data. These are often prone to hallucinations, licensing issues and the generation process is often hard to scale and interpret. In this work, we present a programmatic approach that employs scene graphs as symbolic representations of images and human-written programs to systematically synthesize vision-centric instruction data. Our approach ensures the interpretability and controllability of the data generation process and scales efficiently while maintaining factual accuracy. By implementing a suite of 24 single-image, 14 multi-image instruction generators, and a scene graph generation pipeline, we build a scalable, cost-effective system: ProVision which produces diverse question-answer pairs concerning objects, attributes, relations, depth, etc., for any given image. Applied to Visual Genome and DataComp datasets, we generate over 10 million instruction data points, ProVision-10M, and leverage them in both pretraining and instruction tuning stages of MLMs. When adopted in the instruction tuning stage, our single-image instruction data yields up to a 7% improvement on the 2D split and 8% on the 3D split of CVBench, along with a 3% increase in performance on QBench2, RealWorldQA, and MMMU. Our multi-image instruction data leads to an 8% improvement on Mantis-Eval. Incorporation of our data in both pre-training and fine-tuning stages of xGen-MM-4B leads to an averaged improvement of 1.6% across 11 benchmarks., Comment: code: https://github.com/JieyuZ2/ProVision dataset: https://huggingface.co/datasets/Salesforce/ProVision-10M

While open-source multi-modal language models perform well on simple question answering tasks, they often fail on complex questions that require multiple capabilities, such as fine-grained recognition, visual grounding, and reasoning, and that demand multi-step solutions. We present TACO, a family of multi-modal large action models designed to improve performance on such complex, multi-step, and multi-modal tasks. During inference, TACO produces chains-of-thought-and-action (CoTA), executes intermediate steps by invoking external tools such as OCR, depth estimation and calculator, then integrates both the thoughts and action outputs to produce coherent responses. To train TACO, we create a large dataset of over 1M synthetic CoTA traces generated with GPT-4o and Python programs. We then experiment with various data filtering and mixing techniques and obtain a final subset of 293K high-quality CoTA examples. This dataset enables TACO to learn complex reasoning and action paths, surpassing existing models trained on instruction tuning data with only direct answers. Our model TACO outperforms the instruction-tuned baseline across 8 benchmarks, achieving a 3.6% improvement on average, with gains of up to 15% in MMVet tasks involving OCR, mathematical reasoning, and spatial reasoning. Training on high-quality CoTA traces sets a new standard for complex multi-modal reasoning, highlighting the need for structured, multi-step instruction tuning in advancing open-source mutli-modal models' capabilities.

Evaluating the output of Large Language Models (LLMs) is one of the most critical aspects of building a performant compound AI system. Since the output from LLMs propagate to downstream steps, identifying LLM errors is crucial to system performance. A common task for LLMs in AI systems is tool use. While there are several benchmark environments for evaluating LLMs on this task, they typically only give a success rate without any explanation of the failure cases. To solve this problem, we introduce SpecTool, a new benchmark to identify error patterns in LLM output on tool-use tasks. Our benchmark data set comprises of queries from diverse environments that can be used to test for the presence of seven newly characterized error patterns. Using SPECTOOL , we show that even the most prominent LLMs exhibit these error patterns in their outputs. Researchers can use the analysis and insights from SPECTOOL to guide their error mitigation strategies.

Despite the success of text retrieval in many NLP tasks, code retrieval remains a largely underexplored area. Most text retrieval systems are tailored for natural language queries, often neglecting the specific challenges of retrieving code. This gap leaves existing models unable to effectively capture the diversity of programming languages and tasks across different domains, highlighting the need for more focused research in code retrieval. To address this, we introduce CodeXEmbed, a family of large-scale code embedding models ranging from 400M to 7B parameters. Our novel training pipeline unifies multiple programming languages and transforms various code-related tasks into a common retrieval framework, enhancing model generalizability and retrieval performance. Our 7B model sets a new state-of-the-art (SOTA) in code retrieval, outperforming the previous leading model, Voyage-Code, by over 20% on CoIR benchmark. In addition to excelling in code retrieval, our models demonstrate competitive performance on the widely adopted BeIR text retrieval benchmark, offering versatility across domains. Experimental results demonstrate that improving retrieval performance significantly enhances end-to-end Retrieval-Augmented Generation (RAG) performance for code-related tasks.

Large Language Models (LLMs) are widely adopted for assisting in software development tasks, yet their performance evaluations have narrowly focused on the functional correctness of generated code. Human programmers, however, require LLM-generated code to be not only correct but also optimally efficient. We propose PerfCodeGen, a training-free framework that enhances the performance of LLM-generated code by incorporating feedback based on runtime during test case execution into the self-refinement iterations. With PerfCodeGen, we achieve speedups for a significantly higher proportion of problems compared to using the base LLM with sophisticated prompting techniques. Applied to open language models like Phi-3-mini, PerfCodeGen achieves runtime efficiency comparable to prompting powerful closed models like GPT-4. We achieve state-of-the-art runtime efficiency on benchmarks such as HumanEval, MBPP, and APPS, frequently surpassing the ground truth reference solutions with PerfCodeGen using GPT-3.5 and GPT-4. Additionally, we demonstrate the effectiveness of our approach in enhancing code quality across a range of open LLMs of varying sizes including Phi-3-mini, Llama 3 8B, Mixtral 8x7B, Command R, and Llama 3 70B.

We introduce BLIP3-KALE, a dataset of 218 million image-text pairs that bridges the gap between descriptive synthetic captions and factual web-scale alt-text. KALE augments synthetic dense image captions with web-scale alt-text to generate factually grounded image captions. Our two-stage approach leverages large vision-language models and language models to create knowledge-augmented captions, which are then used to train a specialized VLM for scaling up the dataset. We train vision-language models on KALE and demonstrate improvements on vision-language tasks. Our experiments show the utility of KALE for training more capable and knowledgeable multimodal models. We release the KALE dataset at https://huggingface.co/datasets/Salesforce/blip3-kale

Pre-trained on massive amounts of code and text data, large language models (LLMs) have demonstrated remarkable achievements in performing code generation tasks. With additional execution-based feedback, these models can act as agents with capabilities to self-refine and improve generated code autonomously. However, on challenging coding tasks with extremely large search space, current agentic approaches still struggle with multi-stage planning, generating, and debugging. To address this problem, we propose CodeTree, a framework for LLM agents to efficiently explore the search space in different stages of the code generation process. Specifically, we adopted a unified tree structure to explicitly explore different coding strategies, generate corresponding coding solutions, and subsequently refine the solutions. In each stage, critical decision-making (ranking, termination, expanding) of the exploration process is guided by both the environmental execution-based feedback and LLM-agent-generated feedback. We comprehensively evaluated CodeTree on 7 code generation benchmarks and demonstrated the significant performance gains of CodeTree against strong baselines. Using GPT-4o as the base model, we consistently achieved top results of 95.1 on HumanEval, 98.7 on MBPP, and 43.0 on CodeContests. On the challenging SWEBench benchmark, our approach led to significant performance gains.

Large language models (LLMs) have shown impressive capabilities, but still struggle with complex reasoning tasks requiring multiple steps. While prompt-based methods like Chain-of-Thought (CoT) can improve LLM reasoning at inference time, optimizing reasoning capabilities during training remains challenging. We introduce LaTent Reasoning Optimization (LaTRO), a principled framework that formulates reasoning as sampling from a latent distribution and optimizes it via variational approaches. LaTRO enables LLMs to concurrently improve both their reasoning process and ability to evaluate reasoning quality, without requiring external feedback or reward models. We validate LaTRO through experiments on GSM8K and ARC-Challenge datasets using multiple model architectures. On GSM8K, LaTRO improves zero-shot accuracy by an average of 12.5% over base models and 9.6% over supervised fine-tuning across Phi-3.5-mini, Mistral-7B, and Llama-3.1-8B. Our findings suggest that pre-trained LLMs possess latent reasoning capabilities that can be unlocked and enhanced through our proposed optimization approach in a self-improvement manner. The code of LaTRO is available at \url{https://github.com/SalesforceAIResearch/LaTRO}.

Customer Relationship Management (CRM) systems are vital for modern enterprises, providing a foundation for managing customer interactions and data. Integrating AI agents into CRM systems can automate routine processes and enhance personalized service. However, deploying and evaluating these agents is challenging due to the lack of realistic benchmarks that reflect the complexity of real-world CRM tasks. To address this issue, we introduce CRMArena, a novel benchmark designed to evaluate AI agents on realistic tasks grounded in professional work environments. Following guidance from CRM experts and industry best practices, we designed CRMArena with nine customer service tasks distributed across three personas: service agent, analyst, and manager. The benchmark includes 16 commonly used industrial objects (e.g., account, order, knowledge article, case) with high interconnectivity, along with latent variables (e.g., complaint habits, policy violations) to simulate realistic data distributions. Experimental results reveal that state-of-the-art LLM agents succeed in less than 40% of the tasks with ReAct prompting, and less than 55% even with function-calling abilities. Our findings highlight the need for enhanced agent capabilities in function-calling and rule-following to be deployed in real-world work environments. CRMArena is an open challenge to the community: systems that can reliably complete tasks showcase direct business value in a popular work environment.

The Xenon-Argon Technology (X-ArT) collaboration presents a study on the dynamics of pure and xenon-doped liquid argon (LAr) scintillation. Using two types of silicon photomultipliers sensitive to different wavelength ranges, we identify a long-lived extreme ultraviolet (EUV) radiative component that enhances the light yield. This component is present in both pure and xenon-doped LAr, becoming more pronounced at higher xenon concentrations, where it complements the traditional collisional energy transfer process. To explain this mechanism, we develop a comprehensive model of the Xe-doped LAr scintillation process that integrates both collisional and radiative contributions. Additionally, we investigate how xenon doping affects LAr scintillation light yield and pulse shape discrimination. Finally, we hypothesize that the EUV component may explain the emission of spurious electrons, a known challenge in light dark matter searches using noble liquids. By characterizing the scintillation dynamics in Xe-doped LAr, identifying the long-lived EUV component, and exploring the potential origin of spurious electrons, this work lays the groundwork for optimizing detector performance and advancing the design and sensitivity of future noble liquids particle detectors., Comment: 20 pages, 11 figures

While frontier large language models (LLMs) are capable tool-using agents, current AI systems still operate in a strict turn-based fashion, oblivious to passage of time. This synchronous design forces user queries and tool-use to occur sequentially, preventing the systems from multitasking and reducing interactivity. To address this limitation, we introduce asynchronous AI agents capable of parallel processing and real-time tool-use. Our key contribution is an event-driven finite-state machine architecture for agent execution and prompting, integrated with automatic speech recognition and text-to-speech. Drawing inspiration from the concepts originally developed for real-time operating systems, this work presents both a conceptual framework and practical tools for creating AI agents capable of fluid, multitasking interactions.

We introduce the Principled Reasoning and Acting (PRAct) framework, a novel method for learning and enforcing action principles from trajectory data. Central to our approach is the use of text gradients from a reflection and optimization engine to derive these action principles. To adapt action principles to specific task requirements, we propose a new optimization framework, Reflective Principle Optimization (RPO). After execution, RPO employs a reflector to critique current action principles and an optimizer to update them accordingly. We develop the RPO framework under two scenarios: Reward-RPO, which uses environmental rewards for reflection, and Self-RPO, which conducts self-reflection without external rewards. Additionally, two RPO methods, RPO-Traj and RPO-Batch, is introduced to adapt to different settings. Experimental results across four environments demonstrate that the PRAct agent, leveraging the RPO framework, effectively learns and applies action principles to enhance performance., Comment: Accepted to SIG CoNLL 2024

We present xGen-MM-Vid (BLIP-3-Video): a multimodal language model for videos, particularly designed to efficiently capture temporal information over multiple frames. BLIP-3-Video takes advantage of the 'temporal encoder' in addition to the conventional visual tokenizer, which maps a sequence of tokens over multiple frames into a compact set of visual tokens. This enables BLIP3-Video to use much fewer visual tokens than its competing models (e.g., 32 vs. 4608 tokens). We explore different types of temporal encoders, including learnable spatio-temporal pooling as well as sequential models like Token Turing Machines. We experimentally confirm that BLIP-3-Video obtains video question-answering accuracies comparable to much larger state-of-the-art models (e.g., 34B), while being much smaller (i.e., 4B) and more efficient by using fewer visual tokens. The project website is at https://www.salesforceairesearch.com/opensource/xGen-MM-Vid/index.html

Time series foundation models have demonstrated impressive performance as zero-shot forecasters. However, achieving effectively unified training on time series remains an open challenge. Existing approaches introduce some level of model specialization to account for the highly heterogeneous nature of time series data. For instance, Moirai pursues unified training by employing multiple input/output projection layers, each tailored to handle time series at a specific frequency. Similarly, TimesFM maintains a frequency embedding dictionary for this purpose. We identify two major drawbacks to this human-imposed frequency-level model specialization: (1) Frequency is not a reliable indicator of the underlying patterns in time series. For example, time series with different frequencies can display similar patterns, while those with the same frequency may exhibit varied patterns. (2) Non-stationarity is an inherent property of real-world time series, leading to varied distributions even within a short context window of a single time series. Frequency-level specialization is too coarse-grained to capture this level of diversity. To address these limitations, this paper introduces Moirai-MoE, using a single input/output projection layer while delegating the modeling of diverse time series patterns to the sparse mixture of experts (MoE) within Transformers. With these designs, Moirai-MoE reduces reliance on human-defined heuristics and enables automatic token-level specialization. Extensive experiments on 39 datasets demonstrate the superiority of Moirai-MoE over existing foundation models in both in-distribution and zero-shot scenarios. Furthermore, this study conducts comprehensive model analyses to explore the inner workings of time series MoE foundation models and provides valuable insights for future research.

Time series foundation models excel in zero-shot forecasting, handling diverse tasks without explicit training. However, the advancement of these models has been hindered by the lack of comprehensive benchmarks. To address this gap, we introduce the General Time Series Forecasting Model Evaluation, GIFT-Eval, a pioneering benchmark aimed at promoting evaluation across diverse datasets. GIFT-Eval encompasses 23 datasets over 144,000 time series and 177 million data points, spanning seven domains, 10 frequencies, multivariate inputs, and prediction lengths ranging from short to long-term forecasts. To facilitate the effective pretraining and evaluation of foundation models, we also provide a non-leaking pretraining dataset containing approximately 230 billion data points. Additionally, we provide a comprehensive analysis of 17 baselines, which includes statistical models, deep learning models, and foundation models. We discuss each model in the context of various benchmark characteristics and offer a qualitative analysis that spans both deep learning and foundation models. We believe the insights from this analysis, along with access to this new standard zero-shot time series forecasting benchmark, will guide future developments in time series foundation models. Code, data, and the leaderboard can be found at https://github.com/SalesforceAIResearch/gift-eval .

Retrieval Augmented Generation (RAG), a paradigm that integrates external contextual information with large language models (LLMs) to enhance factual accuracy and relevance, has emerged as a pivotal area in generative AI. The LLMs used in RAG applications are required to faithfully and completely comprehend the provided context and users' questions, avoid hallucination, handle unanswerable, counterfactual or otherwise low-quality and irrelevant contexts, perform complex multi-hop reasoning and produce reliable citations. In this paper, we introduce SFR-RAG, a small LLM that is instruction-tuned with an emphasis on context-grounded generation and hallucination minimization. We also present ContextualBench, a new evaluation framework compiling multiple popular and diverse RAG benchmarks, such as HotpotQA and TriviaQA, with consistent RAG settings to ensure reproducibility and consistency in model assessments. Experimental results demonstrate that our SFR-RAG-9B model outperforms leading baselines such as Command-R+ (104B) and GPT-4o, achieving state-of-the-art results in 3 out of 7 benchmarks in ContextualBench with significantly fewer parameters. The model is also shown to be resilient to alteration in the contextual information and behave appropriately when relevant context is removed. Additionally, the SFR-RAG model maintains competitive performance in general instruction-following tasks and function-calling capabilities., Comment: Technical report

Autonomous agents powered by large language models (LLMs) have attracted significant research interest. However, the open-source community faces many challenges in developing specialized models for agent tasks, driven by the scarcity of high-quality agent datasets and the absence of standard protocols in this area. We introduce and publicly release xLAM, a series of large action models designed for AI agent tasks. The xLAM series includes five models with both dense and mixture-of-expert architectures, ranging from 1B to 8x22B parameters, trained using a scalable, flexible pipeline that unifies, augments, and synthesizes diverse datasets to enhance AI agents' generalizability and performance across varied environments. Our experimental results demonstrate that xLAM consistently delivers exceptional performance across multiple agent ability benchmarks, notably securing the 1st position on the Berkeley Function-Calling Leaderboard, outperforming GPT-4, Claude-3, and many other models in terms of tool use. By releasing the xLAM series, we aim to advance the performance of open-source LLMs for autonomous AI agents, potentially accelerating progress and democratizing access to high-performance models for agent tasks. Models are available at https://huggingface.co/collections/Salesforce/xlam-models-65f00e2a0a63bbcd1c2dade4, Comment: Technical report for the Salesforce xLAM model series

DarkSide-20k (DS-20k) is a dark matter detection experiment under construction at the Laboratori Nazionali del Gran Sasso (LNGS) in Italy. It utilises ~100 t of low radioactivity argon from an underground source (UAr) in its inner detector, with half serving as target in a dual-phase time projection chamber (TPC). The UAr cryogenics system must maintain stable thermodynamic conditions throughout the experiment's lifetime of >10 years. Continuous removal of impurities and radon from the UAr is essential for maximising signal yield and mitigating background. We are developing an efficient and powerful cryogenics system with a gas purification loop with a target circulation rate of 1000 slpm. Central to its design is a condenser operated with liquid nitrogen which is paired with a gas heat exchanger cascade, delivering a combined cooling power of >8 kW. Here we present the design choices in view of the DS-20k requirements, in particular the condenser's working principle and the cooling control, and we show test results obtained with a dedicated benchmarking platform at CERN and LNGS. We find that the thermal efficiency of the recirculation loop, defined in terms of nitrogen consumption per argon flow rate, is 95 % and the pressure in the test cryostat can be maintained within $\pm$(0.1-0.2) mbar. We further detail a 5-day cool-down procedure of the test cryostat, maintaining a cooling rate typically within -2 K/h, as required for the DS-20k inner detector. Additionally, we assess the circuit's flow resistance, and the heat transfer capabilities of two heat exchanger geometries for argon phase change, used to provide gas for recirculation. We conclude by discussing how our findings influence the finalisation of the system design, including necessary modifications to meet requirements and ongoing testing activities., Comment: 45 pages, 24 figures

We present xGen-VideoSyn-1, a text-to-video (T2V) generation model capable of producing realistic scenes from textual descriptions. Building on recent advancements, such as OpenAI's Sora, we explore the latent diffusion model (LDM) architecture and introduce a video variational autoencoder (VidVAE). VidVAE compresses video data both spatially and temporally, significantly reducing the length of visual tokens and the computational demands associated with generating long-sequence videos. To further address the computational costs, we propose a divide-and-merge strategy that maintains temporal consistency across video segments. Our Diffusion Transformer (DiT) model incorporates spatial and temporal self-attention layers, enabling robust generalization across different timeframes and aspect ratios. We have devised a data processing pipeline from the very beginning and collected over 13M high-quality video-text pairs. The pipeline includes multiple steps such as clipping, text detection, motion estimation, aesthetics scoring, and dense captioning based on our in-house video-LLM model. Training the VidVAE and DiT models required approximately 40 and 642 H100 days, respectively. Our model supports over 14-second 720p video generation in an end-to-end way and demonstrates competitive performance against state-of-the-art T2V models., Comment: Accepted by ECCV24 AI4VA

We propose an empirical approach centered on the spectral dynamics of weights -- the behavior of singular values and vectors during optimization -- to unify and clarify several phenomena in deep learning. We identify a consistent bias in optimization across various experiments, from small-scale ``grokking'' to large-scale tasks like image classification with ConvNets, image generation with UNets, speech recognition with LSTMs, and language modeling with Transformers. We also demonstrate that weight decay enhances this bias beyond its role as a norm regularizer, even in practical systems. Moreover, we show that these spectral dynamics distinguish memorizing networks from generalizing ones, offering a novel perspective on this longstanding conundrum. Additionally, we leverage spectral dynamics to explore the emergence of well-performing sparse subnetworks (lottery tickets) and the structure of the loss surface through linear mode connectivity. Our findings suggest that spectral dynamics provide a coherent framework to better understand the behavior of neural networks across diverse settings.

This report introduces xGen-MM (also known as BLIP-3), a framework for developing Large Multimodal Models (LMMs). The framework comprises meticulously curated datasets, a training recipe, model architectures, and a resulting suite of LMMs. xGen-MM, short for xGen-MultiModal, expands the Salesforce xGen initiative on foundation AI models. Our models undergo rigorous evaluation across a range of tasks, including both single and multi-image benchmarks. Our pre-trained base model exhibits strong in-context learning capabilities and the instruction-tuned model demonstrates competitive performance among open-source LMMs with similar model sizes. In addition, we introduce a safety-tuned model with DPO, aiming to mitigate harmful behaviors such as hallucinations and improve safety. We open-source our models, curated large-scale datasets, and our fine-tuning codebase to facilitate further advancements in LMM research. Associated resources will be available on our project page above.

Large language model (LLM) agents have shown great potential in solving real-world software engineering (SWE) problems. The most advanced open-source SWE agent can resolve over 27% of real GitHub issues in SWE-Bench Lite. However, these sophisticated agent frameworks exhibit varying strengths, excelling in certain tasks while underperforming in others. To fully harness the diversity of these agents, we propose DEI (Diversity Empowered Intelligence), a framework that leverages their unique expertise. DEI functions as a meta-module atop existing SWE agent frameworks, managing agent collectives for enhanced problem-solving. Experimental results show that a DEI-guided committee of agents is able to surpass the best individual agent's performance by a large margin. For instance, a group of open-source SWE agents, with a maximum individual resolve rate of 27.3% on SWE-Bench Lite, can achieve a 34.3% resolve rate with DEI, making a 25% improvement and beating most closed-source solutions. Our best-performing group excels with a 55% resolve rate, securing the highest ranking on SWE-Bench Lite. Our findings contribute to the growing body of research on collaborative AI systems and their potential to solve complex software engineering challenges.

We introduce WarpSci, a domain agnostic framework designed to overcome crucial system bottlenecks encountered in the application of reinforcement learning to intricate environments with vast datasets featuring high-dimensional observation or action spaces. Notably, our framework eliminates the need for data transfer between the CPU and GPU, enabling the concurrent execution of thousands of simulations on a single or multiple GPUs. This high data throughput architecture proves particularly advantageous for data-driven scientific research, where intricate environment models are commonly essential.

SOXS (Son Of X-Shooter) will be the new medium-resolution (R 4500 for 1 slit), high-efficiency, wide-band spectrograph for the ESO NTT at La Silla Observatory, Chile. It will be dedicated to the follow-up of any kind of transient events, ensuring fast time, high efficiency, and availability. It consists of a central structure (common path) that supports two spectrographs optimized for the UV-Visible and a Near-Infrared range. Attached to the common path is the Acquisition and Guiding Camera system (AC), equipped with a filter wheel that can provide science-grade imaging and moderate high-speed photometry. The AC Unit was integrated and aligned during the summer months of 2022 and has since been mounted in the NTTs telescope simulator. This work gives an update on the Acquisition Camera Unit status, describes the Image Quality Tests that were performed, and discusses the AC Optical Performance., Comment: SPIE Proceedings Volume 13096, Ground-based and Airborne Instrumentation for Astronomy X; 1309672 (2024)

SOXS (Son Of X-Shooter) is the new single object spectrograph for the ESO New Technology Telescope (NTT) at the La Silla Observatory, able to cover simultaneously both the UV-VIS and NIR bands (350-2000 nm). The instrument is currently in the integration and test phase, approaching the Preliminary Acceptance in Europe (PAE) before shipment to Chile for commissioning. After the assembly and preliminary test of the control electronics at INAF - Astronomical Observatory of Capodimonte (Napoli), the two main control cabinets of SOXS are now hosted in Padova, connected to the real hardware. This contribution describes the final electronic cabinets layout, the control strategy and the different integration phases, waiting for the Preliminary Acceptance in Europe and the installation of the instrument in Chile.

The Son Of X-Shooter (SOXS) will be the specialized facility to observe any transient event with a flexible scheduler at the ESO New Technology Telescope (NTT) at La Silla, Chile. SOXS is a single object spectrograph offering simultaneous spectral coverage in UV-VIS (350-850 nm) and NIR (800-2000 nm) wavelength regimes with an average of R~4500 for a 1arcsec slit. SOXS also has imaging capabilities in the visible wavelength regime. Currently, SOXS is being integrated at the INAF-Astronomical Observatory of Padova. Subsystem- and system-level tests and verification are ongoing to ensure and confirm that every requirement and performance are met. In this paper, we report on the integration and verification of SOXS as the team and the instrument prepare for the Preliminary Acceptance Europe (PAE).

We present the advancements in the development of the scheduler for the Son Of X-shooter instrument at the ESO-NTT 3.58-m telescope in La Silla, Chile. SOXS is designed as a single-object spectroscopic facility and features a high-efficiency spectrograph with two arms covering the spectral range of 350-2000 nm and a mean resolving power of approximately R=4500. It will conduct UV-visible and near-infrared follow-up observations of astrophysical transients, drawing from a broad pool of targets accessible through the streaming services of wide-field telescopes, both current and future, as well as high-energy satellites. The instrument will cater to various scientific objectives within the astrophysical community, each entailing specific requirements for observation planning. SOXS will operate at the European Southern Observatory (ESO) in La Silla, without the presence of astronomers on the mountain. This poses a unique challenge for the scheduling process, demanding a fully automated algorithm that is autonomously interacting with the appropriate databases and the La Silla Weather API, and is capable of presenting the operator not only with an ordered list of optimal targets (in terms of observing constraints) but also with optimal backups in the event of changing weather conditions. This imposes the necessity for a scheduler with rapid-response capabilities without compromising the optimization process, ensuring the high quality of observations and best use of the time at the telescope. We thus developed a new highly available and scalable architecture, implementing API Restful applications like Docker Containers, API Gateway, and Python-based Flask frameworks. We provide an overview of the current state of the scheduler, which is now ready for the approaching on-site testing during Commissioning phase, along with insights into its web interface and preliminary performance tests.

The SOXS spectrograph, designed for the ESO NTT telescope, operates in both the optical (UV-VIS: 350-850 nm) and NIR (800-2000 nm) bands. This article provides an overview of the final tests conducted on the UV-VIS camera system using a telescope simulator. It details the system's performance evaluation, including key metrics such as gain, readout noise, and linearity, and highlights the advancements made in the upgraded acquisition system. The testing process, conducted in the Padua laboratory, involved comprehensive simulations of the telescope environment to ensure the results closely resemble those expected at the ESO-NTT telescope. The successful completion of these tests confirms the system's readiness for deployment to Chile, where it will be installed on the NTT telescope, marking a significant milestone in the SOXS project., Comment: SPIE Proceedings Volume 13096, Ground-based and Airborne Instrumentation for Astronomy X; 130962U (2024)

SOXS (Son Of X-Shooter) is the new ESO instrument that is going to be installed on the 3.58-m New Technology Telescope at the La Silla Observatory. SOXS is a single object spectrograph offering a wide simultaneous spectral coverage from U- to H-band. Although such an instrument may have potentially a large variety of applications, the consortium designed it with a clear science case: it is going to provide the spectroscopic counterparts to the ongoing and upcoming imaging surveys, becoming one of the main follow-up instruments in the Southern hemisphere for the classification and characterization of transients. The NTT+SOXS system is specialized to observe all transients and variable sources discovered by imaging surveys with a flexible schedule maintained by the consortium, based on a remote scheduler which will interface with the observatory software infrastructure. SOXS is realized timely to be highly synergic with transients discovery machines like the Vera C. Rubin Observatory. The instrument has been integrated and tested in Italy, collecting and assembling subsystems coming from all partners spread over six countries in three continents. The first preparatory activities in Chile have been completed at the telescope. This article gives an updated status of the project before the shipping of the instrument to Chile., Comment: SPIE Astronomical Telescopes + Instrumentation, 2024, Yokohama, Japan

The Instrument Control Software of SOXS (Son Of X-Shooter), the forthcoming spectrograph for the ESO New Technology Telescope at the La Silla Observatory, has reached a mature state of development and is approaching the crucial Preliminary Acceptance in Europe phase. Now that all the subsystems have been integrated in the laboratories of the Padova Astronomical Observatory, the team operates for testing purposes with the whole instrument at both engineering and scientific level. These activities will make use of a set of software peculiarities that will be discussed in this contribution. In particular, we focus on the synoptic panel, the co-rotator system special device, on the Active Flexure Compensation system which controls two separate piezo tip-tilt devices., Comment: 6 pages, 3 figures, SPIE conference

Despite the recent proliferation of large language models (LLMs), their training recipes -- model architecture, pre-training data and optimization algorithm -- are often very similar. This naturally raises the question of the similarity among the resulting models. In this paper, we propose a novel setting, imaginary question answering (IQA), to better understand model similarity. In IQA, we ask one model to generate purely imaginary questions (e.g., on completely made-up concepts in physics) and prompt another model to answer. Surprisingly, despite the total fictionality of these questions, all models can answer each other's questions with remarkable success, suggesting a "shared imagination space" in which these models operate during such hallucinations. We conduct a series of investigations into this phenomenon and discuss implications on model homogeneity, hallucination, and computational creativity.

The dual-phase liquid argon time projection chamber is presently one of the leading technologies to search for dark matter particles with masses below 10 GeV/c$^2$. This was demonstrated by the DarkSide-50 experiment with approximately 50 kg of low-radioactivity liquid argon as target material. The next generation experiment DarkSide-20k, currently under construction, will use 1,000 times more argon and is expected to start operation in 2027. Based on the DarkSide-50 experience, here we assess the DarkSide-20k sensitivity to models predicting light dark matter particles, including Weakly Interacting Massive Particles (WIMPs) and sub-GeV/c$^2$ particles interacting with electrons in argon atoms. With one year of data, a sensitivity improvement to dark matter interaction cross-sections by at least one order of magnitude with respect to DarkSide-50 is expected for all these models. A sensitivity to WIMP--nucleon interaction cross-sections below $1\times10^{-42}$ cm$^2$ is achievable for WIMP masses above 800 MeV/c$^2$. With 10 years exposure, the neutrino fog can be reached for WIMP masses around 5 GeV/c$^2$., Comment: 13 pages, 4 figures, supplementary material (4 figures)

The advancement of function-calling agent models requires diverse, reliable, and high-quality datasets. This paper presents APIGen, an automated data generation pipeline designed to synthesize verifiable high-quality datasets for function-calling applications. We leverage APIGen and collect 3,673 executable APIs across 21 different categories to generate diverse function-calling datasets in a scalable and structured manner. Each data in our dataset is verified through three hierarchical stages: format checking, actual function executions, and semantic verification, ensuring its reliability and correctness. We demonstrate that models trained with our curated datasets, even with only 7B parameters, can achieve state-of-the-art performance on the Berkeley Function-Calling Benchmark, outperforming multiple GPT-4 models. Moreover, our 1B model achieves exceptional performance, surpassing GPT-3.5-Turbo and Claude-3 Haiku. We release a dataset containing 60,000 high-quality entries, aiming to advance the field of function-calling agent domains. The dataset is available on Huggingface: https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k and the project homepage: https://apigen-pipeline.github.io/

Large language models (LLMs) for code are typically trained to align with natural language instructions to closely follow their intentions and requirements. However, in many practical scenarios, it becomes increasingly challenging for these models to navigate the intricate boundary between helpfulness and safety, especially against highly complex yet potentially malicious instructions. In this work, we introduce INDICT: a new framework that empowers LLMs with Internal Dialogues of Critiques for both safety and helpfulness guidance. The internal dialogue is a dual cooperative system between a safety-driven critic and a helpfulness-driven critic. Each critic provides analysis against the given task and corresponding generated response, equipped with external knowledge queried through relevant code snippets and tools like web search and code interpreter. We engage the dual critic system in both code generation stage as well as code execution stage, providing preemptive and post-hoc guidance respectively to LLMs. We evaluated INDICT on 8 diverse tasks across 8 programming languages from 5 benchmarks, using LLMs from 7B to 70B parameters. We observed that our approach can provide an advanced level of critiques of both safety and helpfulness analysis, significantly improving the quality of output codes ($+10\%$ absolute improvements in all models)., Comment: Accepted to The Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024)

Multimodal interleaved datasets featuring free-form interleaved sequences of images and text are crucial for training frontier large multimodal models (LMMs). Despite the rapid progression of open-source LMMs, there remains a pronounced scarcity of large-scale, diverse open-source multimodal interleaved datasets. In response, we introduce MINT-1T, the most extensive and diverse open-source Multimodal INTerleaved dataset to date. MINT-1T comprises one trillion text tokens and 3.4 billion images, a 10x scale-up from existing open-source datasets. Additionally, we include previously untapped sources such as PDFs and ArXiv papers. As scaling multimodal interleaved datasets requires substantial engineering effort, sharing the data curation process and releasing the dataset greatly benefits the community. Our experiments show that LMMs trained on MINT-1T rival the performance of models trained on the previous leading dataset, OBELICS. Our data and code will be released at https://github.com/mlfoundations/MINT-1T.

The deployment of Large Language Models (LLMs) and Large Multimodal Models (LMMs) on mobile devices has gained significant attention due to the benefits of enhanced privacy, stability, and personalization. However, the hardware constraints of mobile devices necessitate the use of models with fewer parameters and model compression techniques like quantization. Currently, there is limited understanding of quantization's impact on various task performances, including LLM tasks, LMM tasks, and, critically, trust and safety. There is a lack of adequate tools for systematically testing these models on mobile devices. To address these gaps, we introduce MobileAIBench, a comprehensive benchmarking framework for evaluating mobile-optimized LLMs and LMMs. MobileAIBench assesses models across different sizes, quantization levels, and tasks, measuring latency and resource consumption on real devices. Our two-part open-source framework includes a library for running evaluations on desktops and an iOS app for on-device latency and hardware utilization measurements. Our thorough analysis aims to accelerate mobile AI research and deployment by providing insights into the performance and feasibility of deploying LLMs and LMMs on mobile platforms.

Experiments aimed at direct searches for WIMP dark matter require highly effective reduction of backgrounds and control of any residual radioactive contamination. In particular, neutrons interacting with atomic nuclei represent an important class of backgrounds due to the expected similarity of a WIMP-nucleon interaction, so that such experiments often feature a dedicated neutron detector surrounding the active target volume. In the context of the development of DarkSide-20k detector at INFN Gran Sasso National Laboratory (LNGS), several R&D projects were conceived and developed for the creation of a new hybrid material rich in both hydrogen and gadolinium nuclei to be employed as an essential element of the neutron detector. Thanks to its very high cross-section for neutron capture, gadolinium is one of the most widely used elements in neutron detectors, while the hydrogen-rich material is instrumental in efficiently moderating the neutrons. In this paper results from one of the R&Ds are presented. In this effort the new hybrid material was obtained as a poly(methyl methacrylate) (PMMA) matrix, loaded with gadolinium oxide in the form of nanoparticles. We describe its realization, including all phases of design, purification, construction, characterization, and determination of mechanical properties of the new material.

Resistive random-access memory (RRAM) is widely recognized as a promising emerging hardware platform for deep neural networks (DNNs). Yet, due to manufacturing limitations, current RRAM devices are highly susceptible to hardware defects, which poses a significant challenge to their practical applicability. In this paper, we present a machine learning technique that enables the deployment of defect-prone RRAM accelerators for DNN applications, without necessitating modifying the hardware, retraining of the neural network, or implementing additional detection circuitry/logic. The key idea involves incorporating a drop-connect inspired approach during the training phase of a DNN, where random subsets of weights are selected to emulate fault effects (e.g., set to zero to mimic stuck-at-1 faults), thereby equipping the DNN with the ability to learn and adapt to RRAM defects with the corresponding fault rates. Our results demonstrate the viability of the drop-connect approach, coupled with various algorithm and system-level design and trade-off considerations. We show that, even in the presence of high defect rates (e.g., up to 30%), the degradation of DNN accuracy can be as low as less than 1% compared to that of the fault-free version, while incurring minimal system-level runtime/energy costs.

Autonomous agents that accomplish complex computer tasks with minimal human interventions have the potential to transform human-computer interaction, significantly enhancing accessibility and productivity. However, existing benchmarks either lack an interactive environment or are limited to environments specific to certain applications or domains, failing to reflect the diverse and complex nature of real-world computer use, thereby limiting the scope of tasks and agent scalability. To address this issue, we introduce OSWorld, the first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across various operating systems such as Ubuntu, Windows, and macOS. OSWorld can serve as a unified, integrated computer environment for assessing open-ended computer tasks that involve arbitrary applications. Building upon OSWorld, we create a benchmark of 369 computer tasks involving real web and desktop apps in open domains, OS file I/O, and workflows spanning multiple applications. Each task example is derived from real-world computer use cases and includes a detailed initial state setup configuration and a custom execution-based evaluation script for reliable, reproducible evaluation. Extensive evaluation of state-of-the-art LLM/VLM-based agents on OSWorld reveals significant deficiencies in their ability to serve as computer assistants. While humans can accomplish over 72.36% of the tasks, the best model achieves only 12.24% success, primarily struggling with GUI grounding and operational knowledge. Comprehensive analysis using OSWorld provides valuable insights for developing multimodal generalist agents that were not possible with previous benchmarks. Our code, environment, baseline models, and data are publicly available at https://os-world.github.io., Comment: 51 pages, 21 figures

We present BEHAVIOR-1K, a comprehensive simulation benchmark for human-centered robotics. BEHAVIOR-1K includes two components, guided and motivated by the results of an extensive survey on "what do you want robots to do for you?". The first is the definition of 1,000 everyday activities, grounded in 50 scenes (houses, gardens, restaurants, offices, etc.) with more than 9,000 objects annotated with rich physical and semantic properties. The second is OMNIGIBSON, a novel simulation environment that supports these activities via realistic physics simulation and rendering of rigid bodies, deformable bodies, and liquids. Our experiments indicate that the activities in BEHAVIOR-1K are long-horizon and dependent on complex manipulation skills, both of which remain a challenge for even state-of-the-art robot learning solutions. To calibrate the simulation-to-reality gap of BEHAVIOR-1K, we provide an initial study on transferring solutions learned with a mobile manipulator in a simulated apartment to its real-world counterpart. We hope that BEHAVIOR-1K's human-grounded nature, diversity, and realism make it valuable for embodied AI and robot learning research. Project website: https://behavior.stanford.edu., Comment: A preliminary version was published at 6th Conference on Robot Learning (CoRL 2022)

Monochromatic gamma-ray signals constitute a potential smoking gun signature for annihilating or decaying dark matter particles that could relatively easily be distinguished from astrophysical or instrumental backgrounds. We provide an updated assessment of the sensitivity of the Cherenkov Telescope Array (CTA) to such signals, based on observations of the Galactic centre region as well as of selected dwarf spheroidal galaxies. We find that current limits and detection prospects for dark matter masses above 300 GeV will be significantly improved, by up to an order of magnitude in the multi-TeV range. This demonstrates that CTA will set a new standard for gamma-ray astronomy also in this respect, as the world's largest and most sensitive high-energy gamma-ray observatory, in particular due to its exquisite energy resolution at TeV energies and the adopted observational strategy focussing on regions with large dark matter densities. Throughout our analysis, we use up-to-date instrument response functions, and we thoroughly model the effect of instrumental systematic uncertainties in our statistical treatment. We further present results for other potential signatures with sharp spectral features, e.g.~box-shaped spectra, that would likewise very clearly point to a particle dark matter origin., Comment: 44 pages JCAP style (excluding author list and references), 19 figures; minor changes to match published version

The booming success of LLMs initiates rapid development in LLM agents. Though the foundation of an LLM agent is the generative model, it is critical to devise the optimal reasoning strategies and agent architectures. Accordingly, LLM agent research advances from the simple chain-of-thought prompting to more complex ReAct and Reflection reasoning strategy; agent architecture also evolves from single agent generation to multi-agent conversation, as well as multi-LLM multi-agent group chat. However, with the existing intricate frameworks and libraries, creating and evaluating new reasoning strategies and agent architectures has become a complex challenge, which hinders research investigation into LLM agents. Thus, we open-source a new AI agent library, AgentLite, which simplifies this process by offering a lightweight, user-friendly platform for innovating LLM agent reasoning, architectures, and applications with ease. AgentLite is a task-oriented framework designed to enhance the ability of agents to break down tasks and facilitate the development of multi-agent systems. Furthermore, we introduce multiple practical applications developed with AgentLite to demonstrate its convenience and flexibility. Get started now at: \url{https://github.com/SalesforceAIResearch/AgentLite}., Comment: preprint. Library is available at https://github.com/SalesforceAIResearch/AgentLite