74 accepted papers
Scaling up high-quality trajectory data has long been a critical bottleneck for developing human-like computer use agents. We introduce PC Agent-E, an efficient agent training framework that significantly reduces reliance on large-scale human demonstrations. Starting with just 312 human-annotated co…
Hand-craft reward engineering requires domain knowledge with numerous trials and errors, while Preference-based Reinforcement Learning (PbRL) avoids manual reward design but often suffers from limited interpretability and unstable training. To address these issues, we propose a novel preference alig
Knowledge distillation (KD) is a powerful paradigm for compressing large language models (LLMs), whose effectiveness depends on intertwined choices of divergence direction, optimization strategy, and data regime. We break down the design of existing KD methods and present a unified view that establi…
AI agents could accelerate scientific discovery by automating hypothesis formation, experiment design, coding, execution, and analysis, yet existing benchmarks probe narrow skills in simplified settings. To address this gap, we introduce InnovatorBench, a benchmark-platform pair for realistic, end-t…
Diffusion Large Language Models (dLLMs) enable parallel token generation, and their block-wise variants have attracted significant attention. However, existing dLLMs usually exhibit an accuracy–parallelism trade-off, where raising tokens per forward (TPF) via aggressive parallel decoding often degra…
Recent advances in Vision-Language-Action (VLA) models demonstrate that visual signals can effectively complement sparse action supervisions. However, letting VLA directly predict high-dimensional visual states can distribute model capacity and incur prohibitive training cost, while compressing visu
Supervised fine-tuning (SFT) of foundation models often leads to poor generalization, where prior capabilities deteriorate after tuning on specific tasks. Inspired by trust-region policy optimization (TRPO) and proximal policy optimization (PPO) in reinforcement learning (RL), we propose Proximal SF…
Test-time compute scaling has emerged as a powerful paradigm for enhancing mathematical reasoning in large language models (LLMs) by allocating additional computational resources during inference. However, current methods employ uniform resource distribution across all reasoning sub-problems, creati
Recently, Large Language Models (LLMs) have been applied to scientific equation discovery, leveraging their embedded scientific knowledge for hypothesis generation. However, current methods typically confine LLMs to the role of an equation proposer within search algorithms like genetic programming.…
As Large Language Models (LLMs) are increasingly deployed in highly specialized vertical domains, the evaluation of their domain-specific performance becomes critical. However, existing evaluations for vertical domains typically rely on the labor-intensive construction of static single-turn datasets…
Vision tool-use reinforcement learning (RL) can equip vision--language models with visual operators such as crop-and-zoom and achieves strong performance gains, yet it remains unclear whether these gains are driven by improvements in tool use or evolving intrinsic capabilities. We introduce \textbf{…
Recently, the frontier of Large Language Model (LLM) capabilities has shifted from single-turn code generation to agentic software engineering—a paradigm where models autonomously navigate, edit, and test complex repositories. While post-training methods have become the de facto approach for code ag…
We introduce DavIR, a model-based data selection method for post-training Large Language Models. DavIR generalizes Reducible Holdout Loss to core-set selection problem of causal language modeling, and quantifies the learnability of a given datum with respect to a pre-trained LLM based on relative re…
Large Language Models (LLMs) with web search capabilities show significant potential for deep research, yet current methods—brittle prompt engineering or RAG-based reinforcement learning in controlled environments—fail to capture real-world complexities. In this paper, we introduce DeepResearcher, t
The leaderboard of Large Language Models (LLMs) in mathematical tasks has been continuously updated. However, the majority of evaluations focus solely on the final results, neglecting the quality of the intermediate steps. This oversight can mask underlying problems, such as logical errors or unnece…
Large language models (LLMs) have demonstrated remarkable reasoning capabilities through test-time scaling approaches, particularly when fine-tuned with chain-of-thought (CoT) data distilled from more powerful large reasoning models (LRMs). However, these reasoning chains often contain verbose eleme…
As large language models (LLMs) continue to evolve, leaderboards play a significant role in steering their development. Existing leaderboards often prioritize model capabilities while overlooking safety concerns, leaving a significant gap in responsible AI development. To address this gap, we introd…
Vision-language instruction-tuning models have recently achieved significant performance improvements. In this work, we discover that large-scale 3D parallel training on those models leads to an imbalanced computation load across different devices. The vision and language parts are inherently hetero…
Alignment with human preferences is an important evaluation aspect of LLMs, requiring them to be helpful, honest, safe, and to precisely follow human instructions. Evaluating large language models' (LLMs) alignment typically involves directly assessing their open-ended responses, requiring human ann…
Large language model pre-training has traditionally relied on human experts to craft heuristics for improving the corpora quality, resulting in numerous rules developed to date. However, these fixed rules lack the flexibility to address the unique characteristics of individual examples, yet crafting…
Self-improvement through post-training methods such as iterative preference learning has been acclaimed for enhancing the problem-solving capabilities (e.g., mathematical reasoning) of Large Language Models (LLMs) without human intervention. However, as our exploration deepens, it is crucial to crit…
Although diffusion models can generate high-quality human images, their applications are limited by the instability in generating hands with correct structures. In this paper, we introduce RHanDS, a conditional diffusion-based framework designed to refine malformed hands by utilizing decoupled struc…
As Large Language Models (LLMs) increasingly participate in human-AI interactions, evaluating their Theory of Mind (ToM) capabilities - particularly their ability to track dynamic mental states - becomes crucial. While existing benchmarks assess basic ToM abilities, they predominantly focus on stati…
Direct Preference Optimization (DPO) has become a widely used training method for the instruction fine-tuning of large language models (LLMs). In this work, we explore an under-investigated aspect of DPO – its dependency on the reference model or policy. Such reference policies, typically instantiat…
Aligning language models (LMs) with human preferences has become a key area of research, enabling these models to meet diverse user needs better. Inspired by weak-to-strong generalization, where a strong LM fine-tuned on labels generated by a weaker model can consistently outperform its weak supervi…
Recent research has made significant strides in aligning large language models (LLMs) with helpfulness and harmlessness. In this paper, we argue for the importance of alignment for \emph{honesty}, ensuring that LLMs proactively refuse to answer questions when they lack knowledge, while still not bei…
While large language models (LLMs) can already achieve strong performance on standard generic summarization benchmarks, their performance on more complex summarization task settings is less studied. Therefore, we benchmark LLMs on instruction controllable text summarization, where the model input co…
As a relative quality comparison of model responses, human and Large Language Model (LLM) preferences serve as common alignment goals in model fine-tuning and criteria in evaluation. Yet, these preferences merely reflect broad tendencies, resulting in less explainable and controllable models with po…
Medical text generation aims to assist with administrative work and highlight salient information to support decision-making.To reflect the specific requirements of medical text, in this paper, we propose a set of metrics to evaluate the completeness, conciseness, and attribution of the generated te…
The rise of large language models (LLMs) has significantly influenced the quality of information in decision-making systems, leading to the prevalence of AI-generated content and challenges in detecting misinformation and managing conflicting information, or “inter-evidence conflicts.” This study in…
Fuzzy reasoning is vital due to the frequent use of imprecise information in daily contexts. However, the ability of current large language models (LLMs) to handle such reasoning remains largely uncharted. In this paper, we introduce a new benchmark, FRoG, for fuzzy reasoning, featuring real-world m…
Generative Artificial Intelligence (AI) has enabled the development of sophisticated models that are capable of producing high-caliber text, images, and other outputs through the utilization of large pre-trained models.Nevertheless, assessing the quality of the generation is an even more arduous tas…
The rapid development of Large Language Models (LLMs) has substantially expanded the range of tasks they can address. In the field of Natural Language Processing (NLP), researchers have shifted their focus from conventional NLP tasks (e.g., sequence tagging and parsing) towards tasks that revolve ar…
This paper introduces the Decomposed Requirements Following Ratio (DRFR), a new metric for evaluating Large Language Models’ (LLMs) ability to follow instructions. Addressing a gap in current methodologies, DRFR breaks down complex instructions into simpler criteria, facilitating a detailed analysis…
Large Language Models (LLMs) have shown impressive capabilities but also a concerning tendency to hallucinate. This paper presents RefChecker, a framework that introduces claim-triplets to represent claims in LLM responses, aiming to detect fine-grained hallucinations. In RefChecker, an extractor ge…
Humans follow criteria when they execute tasks, and these criteria are directly used to assess the quality of task completion. Therefore, having models learn to use criteria to provide feedback can help humans or models to perform tasks better. However, current research in this area tends to conside…
High-quality, large-scale corpora are the cornerstone of building foundation models. In this work, we introduce MathPile, a diverse and high-quality math-centric corpus comprising about 9.5 billion tokens. Throughout its creation, we adhered to the principle of “less is more”, firmly believing in th…
A story premise succinctly defines a story’s main idea, foundation, and trajectory. It serves as the initial trigger in automatic story generation. Existing sources of story premises are limited by a lack of diversity, uneven quality, and high costs that make them difficult to scale. In response, we…
The evolution of Artificial Intelligence (AI) has been significantly accelerated by advancements in Large Language Models (LLMs) and Large Multimodal Models (LMMs), gradually showcasing potential cognitive reasoning abilities in problem-solving and scientific discovery (i.e., AI4Science) once exclus…
Recent studies have found that summaries generated by large language models (LLMs) are favored by human annotators over the original reference summaries in commonly used summarization datasets. Therefore, we study an LLM-as-reference learning setting for smaller text summarization models to investig…
The rapid growth of scientific literature imposes significant challenges for researchers endeavoring to stay updated with the latest advancements in their fields and delve into new areas. We introduce OpenResearcher, an innovative platform that leverages Artificial Intelligence (AI) techniques to ac…
Despite Retrieval-Augmented Generation (RAG) has shown promising capability in leveraging external knowledge, a comprehensive evaluation of RAG systems is still challenging due to the modular nature of RAG, evaluation of long-form responses and reliability of measurements. In this paper, we propose…
The quality of finetuning data is crucial for aligning large language models (LLMs) with human values. Current methods to improve data quality are either labor-intensive or prone to factual errors caused by LLM hallucinations. This paper explores elevating the quality of existing instruction data to…
The deployment of Large Language Models (LLMs) in content generation raises significant safety concerns, particularly regarding the transparency and interpretability of content evaluations. Current methods, primarily focused on binary safety classifications, lack mechanisms for detailed critique, li…
When large language models (LLMs) surpass human capabilities, supervising them effectively becomes difficult. Weak-to-strong learning, where a less capable model enhances a stronger one, proves valuable in this context. Yet, the efficacy of this paradigm for complex reasoning tasks is still unexplor…
Modern machine learning relies on datasets to develop and validate research ideas. Given the growth of publicly available data, finding the right dataset to use is increasingly difficult. Any research question imposes explicit and implicit constraints on how well a given dataset will enable research…
Assessing factuality of text generated by large language models (LLMs) is an emerging yet crucial research area, aimed at alerting users to potential errors and guiding the development of more reliable LLMs. Nonetheless, the evaluators assessing factuality necessitate suitable evaluation themselves…
Despite the major advances in NLP, significant disparities in NLP system performance across languages still exist. Arguably, these are due to uneven resource allocation and sub-optimal incentives to work on less resourced languages. To track and further incentivize the global development of equitabl…
Large language models are trained in two stages: (1) unsupervised pretraining from raw text, to learn general-purpose representations, and (2) large scale instruction tuning and reinforcement learning, to better align to end tasks and user preferences. We measure the relative importance of these tw…
Evaluation of natural language generation (NLG) is complex and multi-dimensional. Generated text can be evaluated for fluency, coherence, factuality, or any other dimensions of interest. Most frameworks that perform such multi-dimensional evaluation require training on large manually or syntheticall…
Large language models (LLMs) have demonstrated an impressive ability to perform arithmetic and symbolic reasoning tasks, when provided with a few examples at test time ("few-shot prompting"). Much of this success can be attributed to prompting methods such as "chain-of-thought", which employ LLMs fo…
Human evaluation is the foundation upon which the evaluation of both summarization systems and automatic metrics rests. However, existing human evaluation studies for summarization either exhibit a low inter-annotator agreement or have insufficient scale, and an in-depth analysis of human evaluation…
Modern embedding-based metrics for evaluation of generated text generally fall into one of two paradigms: discriminative metrics that are trained to directly predict which outputs are of higher quality according to supervised human annotations, and generative metrics that are trained to evaluate tex…
In this paper, we ask the research question of whether all the datasets in the benchmark are necessary. We approach this by first characterizing the distinguishability of datasets when comparing different systems. Experiments on 9 datasets and 36 systems show that several existing benchmark datasets…
Abstractive summarization models are commonly trained using maximum likelihood estimation, which assumes a deterministic (one-point) target distribution in which an ideal model will assign all the probability mass to the reference summary. This assumption may lead to performance degradation during i…
Measuring the fidelity of synthesized virtual traffic has become an important and fundamental concern for evaluating the performance of different traffic simulation techniques and applications of autonomous vehicle testing. In this work, we propose a novel method to evaluate the fidelity of any traj…
In this paper, we present the Intra- and Inter-Human Relation Networks I²R-Net for Multi-Person Pose Estimation. It involves two basic modules. First, the Intra-Human Relation Module operates on a single person and aims to capture Intra-Human dependencies. Second, the Inter-Human Relation Module con…
The surge in the number of scientific submissions has brought challenges to the work of peer review. In this paper, as a first step, we explore the possibility of designing an automated system, which is not meant to replace humans, but rather providing a first-pass draft for a machine-assisted human…
This paper aims for a potential architectural improvement for multilingual learning and asks: Can different tasks from different languages be modeled in a monolithic framework, i.e. without any task/language-specific module? The benefit of achieving this could open new doors for future multilingual…
Multi-dimensional evaluation is the dominant paradigm for human evaluation in Natural Language Generation (NLG), i.e., evaluating the generated text from multiple explainable dimensions, such as coherence and fluency. However, automatic evaluation in NLG is still dominated by similarity-based metric…
With the continuous upgrading of the summarization systems driven by deep neural networks, researchers have higher requirements on the quality of the generated summaries, which should be not only fluent and informative but also factually correct. As a result, the field of factual evaluation has deve…
A wide variety of NLP applications, such as machine translation, summarization, and dialog, involve text generation. One major challenge for these applications is how to evaluate whether such generated texts are actually fluent, accurate, or effective. In this work, we conceptualize the evaluation o…
Automatically extracting key information from scientific documents has the potential to help scientists work more efficiently and accelerate the pace of scientific progress. Prior work has considered extracting document-level entity clusters and relations end-to-end from raw scientific text, which c…
Aspect-based Sentiment Analysis (ABSA), aiming at predicting the polarities for aspects, is a fine-grained task in the field of sentiment analysis. Previous work showed syntactic information, e.g. dependency trees, can effectively improve the ABSA performance. Recently, pre-trained models (PTMs) als…
Neural abstractive summarization models are flexible and can produce coherent summaries, but they are sometimes unfaithful and can be difficult to control. While previous studies attempt to provide different types of guidance to control the output and increase faithfulness, it is not clear how these…
The development of neural networks and pretraining techniques has spawned many sentence-level tagging systems that achieved superior performance on typical benchmarks. However, a relatively less discussed topic is what if more context information is introduced into current top-scoring tagging system…
Although some recent works show potential complementarity among different state-of-the-art systems, few works try to investigate this problem in text summarization. Researchers in other areas commonly refer to the techniques of reranking or stacking to approach this problem. In this work, we highlig…
In this paper, we present a conceptually simple while empirically powerful framework for abstractive summarization, SimCLS, which can bridge the gap between the learning objective and evaluation metrics resulting from the currently dominated sequence-to-sequence learning framework by formulating tex…
Recent years have seen the paradigm shift of Named Entity Recognition (NER) systems from sequence labeling to span prediction. Despite its preliminary effectiveness, the span prediction model’s architectural bias has not been fully understood. In this paper, we first investigate the strengths and we…
Machine learning has brought striking advances in multilingual natural language processing capabilities over the past year. For example, the latest techniques have improved the state-of-the-art performance on the XTREME multilingual benchmark by more than 13 points. While a sizeable gap to human-lev…
In text summarization, evaluating the efficacy of automatic metrics without human judgments has become recently popular. One exemplar work (Peyrard, 2019) concludes that automatic metrics strongly disagree when ranking high-scoring summaries. In this paper, we revisit their experiments and find that…
In this work, we propose an efficient and accurate monocular 3D detection framework in single shot. Most successful 3D detectors take the projection constraint from the 3D bounding box to the 2D box as an important component. Four edges of a 2D box provide only four constraints and the performance d…
Region-based Fully ConvNet (R-FCN) designed for general object detection is difficult to be directly applied for pedestrian detection, due to being with large human pose and scale changes, and even with partial occlusion in surveillance scenarios. This paper presents a real time pedestrian detection…
This paper proposed a novel method for face hallucination based on a neighbor embedding technique. Traditional neighbor embedding approaches often offer counterintuitive results because consistency between high resolution images and low resolution images cannot be preserved without taking the intrin…