Reinforcement learning (RL) has become a central paradigm for post-training large language models (LLMs), particularly for complex reasoning tasks, yet it often suffers from exploration collapse: policies prematurely concentrate on a small set of dominant reasoning patterns, improving pass@1 while limiting rollout-level diversity and gains in pass@k. We argue that this failure stems from regularizing local token behavior rather than diversity over sets of solutions. To address this, we propose Uniqueness-Aware Reinforcement Learning, a rollout-level objective that explicitly rewards correct solutions that exhibit rare high-level strategies. Our method uses an LLM-based judge to cluster rollouts for the same problem according to their high-level solution strategies, ignoring superficial variations, and reweights policy advantages inversely with cluster size. As a result, correct but novel strategies receive higher rewards than redundant ones. Across mathematics, physics, and medical reasoning benchmarks, our approach consistently improves pass@k across large sampling budgets and increases the area under the pass@k curve (AUC@K) without sacrificing pass@1, while sustaining exploration and uncovering more diverse solution strategies at scale. Code is in Software part under submission page.

Large reasoning models (LRMs) already possess a latent capacity for long chain-of-thought reasoning. Prior work has shown that outcome-based reinforcement learning (RL) can incidentally elicit advanced reasoning behaviors such as self-correction, backtracking, and verification–phenomena often referred to as the model’s ”aha moment”. However, the timing and consistency of these emergent behaviors remain unpredictable and uncontrollable, limiting the scalability and reliability of LRMs’ reasoning capabilities. To address these limitations, we move beyond reliance on prompts and unpredictable ”aha moments”. Instead, we explicitly align models with three meta-abilities: deduction, induction, and abduction, using automatically generated, self-verifiable tasks. Our three-stage pipeline (individual alignment, parameter-space merging, domain-specific reinforcement learning) boosts performance by over 10% relative to instruction-tuned baselines. Furthermore, domain-specific RL from the aligned checkpoint yields an additional gain in performance ceiling for both 7B and 32B models across math, coding, and science benchmarks, showing that explicit meta-ability alignment offers a scalable and dependable foundation for reasoning. Code and data can be found in Software and Data part in submission page.

Large Language Models (LLMs) excel in stand-alone code tasks like HumanEval and MBPP, but struggle with handling entire code repositories. This challenge has prompted research on enhancing LLM-codebase interaction at a repository scale. Current solutions rely on similarity-based retrieval or manual tools and APIs, each with notable drawbacks. Similarity-based retrieval often has low recall in complex tasks, while manual tools and APIs are typically task-specific and require expert knowledge, reducing their generalizability across diverse code tasks and real-world applications. To mitigate these limitations, we introduce CodexGraph, a system that integrates LLM agents with graph database interfaces extracted from code repositories. By leveraging the structural properties of graph databases and the flexibility of the graph query language, CodexGraph enables the LLM agent to construct and execute queries, allowing for precise, code structure-aware context retrieval and code navigation. We assess CodexGraph using three benchmarks: CrossCodeEval, SWE-bench, and EvoCodeBench. Additionally, we develop five real-world coding applications. With a unified graph database schema, CodexGraph demonstrates competitive performance and potential in both academic and real-world environments, showcasing its versatility and efficacy in software engineering. Our code and demo will be released soon.

Large Language Models (LLMs) as agents require careful behavioral adaptation. While adept at reactive tasks (e.g., medical reasoning), LLMs often struggle with proactive engagement, like unprompted identification of critical missing information or risks. We introduce **BehaviorBench**, a comprehensive dataset to evaluate agent behaviors across a clinical assistance spectrum. To rigorously test the current models, we also introduce **BehaviorBench-Hard**, a challenging subset where the performance of state-of-the-art models drops significantly, revealing weaknesses. To address these challenges, we propose **BehaviorSFT**, a novel training strategy using behavioral tokens to explicitly condition LLMs for dynamic behavioral selection which boosts performance on both benchmarks. Crucially, a blind clinician evaluation confirmed that our trained agents exhibit more realistic clinical behavior, striking a superior balance between helpful proactivity and necessary restraint versus standard fine-tuning or explicitly instructed agents. Project Page: https://behavior-adaptation.github.io/

Large language models (LLMs) face significant challenges in handling long-context tasks because of their limited effective context window size during pretraining, which restricts their ability to generalize over extended sequences. Meanwhile, extending the context window in LLMs through post-pretraining is highly resource-intensive.To address this, we introduce LongRecipe, an efficient training strategy for extending the context window of LLMs, including impactful token analysis, position index transformation, and training optimization strategies. It simulates long-sequence inputs while maintaining training efficiency and significantly improves the model’s understanding of long-range dependencies. Experiments on three types of LLMs show that LongRecipe can utilize long sequences while requiring only 30% of the target context window size, and reduces computational training resource over 85% compared to full sequence training. Furthermore, LongRecipe also preserves the original LLM’s capabilities in general tasks. Ultimately, we can extend effective context window of open-source LLMs from 8k to 128k, achieving performance close to GPT-4 with just one day of dedicated training using a single GPU with 80G memory.Our code is released at https://github.com/zhiyuanhubj/LongRecipe.

Large Language Models (LLMs) have significantly advanced natural language processing, demonstrating exceptional reasoning, tool usage, and memory capabilities. As their applications expand into multi-agent environments, there arises a need for a comprehensive evaluation framework that captures LLMs’ reasoning, planning, collaboration, and other social abilities. This work introduces a novel competition-based benchmark framework specifically designed to assess LLMs within multi-agent settings, providing quantitative metrics to evaluate their judgment, reasoning, deception, self-awareness, cooperation, coordination, and rationality.We utilize two social deduction games alongside three game-theory scenarios to create diverse environments.Our frame is fortified with the probabilistic graphic modeling (PGM) method, enhancing the LLMs’ capabilities in navigating complex social and cognitive dimensions. We evaluate seven LLMs, quantitatively highlighting a significant capability gap of over threefold between the strongest, GPT o1, and the weakest, Llama-2-70B. It also confirms that our PGM enhancement boosts the abilities of all selected models by an average of 37%. Our data and code can be found here https://github.com/cathyxl/MAgIC.

Seeking answers effectively for long videos is essential to build video question answering (videoQA) systems. Previous methods adaptively select frames and regions from long videos to save computations. However, this fails to reason over the whole sequence of video, leading to sub-optimal performance. To address this problem, we introduce a state space layer (SSL) into multi-modal Transformer to efficiently integrate global semantics of the video, which mitigates the video information loss caused by frame and region selection modules. Our SSL includes a gating unit to enable controllability over the flow of global semantics into visual representations. To further enhance the controllability, we introduce a cross-modal compositional congruence objective to encourage global semantics aligned with the question. To rigorously evaluate long-form videoQA capacity, we construct two new benchmarks Ego-QA and MAD-QA featuring videos of considerably long length, i.e. 17.5 minutes and 1.9 hours, respectively. Extensive experiments demonstrate the superiority of our framework on these new as well as existing datasets.

Dialogue systems have received increasing attention while automatically evaluating their performance remains challenging. User satisfaction estimation (USE) has been proposed as an alternative. It assumes that the performance of a dialogue system can be measured by user satisfaction and uses an estimator to simulate users. The effectiveness of USE depends heavily on the estimator. Existing estimators independently predict user satisfaction at each turn and ignore satisfaction dynamics across turns within a dialogue. In order to fully simulate users, it is crucial to take satisfaction dynamics into account. To fill this gap, we propose a new estimator ASAP (sAtisfaction eStimation via HAwkes Process) that treats user satisfaction across turns as an event sequence and employs a Hawkes process to effectively model the dynamics in this sequence. Experimental results on four benchmark dialogue datasets demonstrate that ASAP can substantially outperform state-of-the-art baseline estimators.

We propose a novel strategy to encode the syntax parse tree of sentence into a learnable distributed representation. The proposed syntax encoding scheme is provably information-lossless. In specific, an embedding vector is constructed for each word in the sentence, encoding the path in the syntax tree corresponding to the word. The one-to-one correspondence between these “syntax-embedding” vectors and the words (hence their embedding vectors) in the sentence makes it easy to integrate such a representation with all word-level NLP models. We empirically show the benefits of the syntax embeddings on the Authorship Attribution domain, where our approach improves upon the prior art and achieves new performance records on five benchmarking data sets.