English
Related papers

Related papers: AMR-SD: Asymmetric Meta-Reflective Self-Distillati…

200 papers

Large Language Models (LLMs) empowered with Tool-Integrated Reasoning (TIR) can iteratively plan, call external tools, and integrate returned information to solve complex, long-horizon reasoning tasks. Agentic Reinforcement Learning…

Computation and Language · Computer Science 2026-01-21 Jianghao Su , Xia Zeng , Luhui Liu , Chao Luo , Ye Chen , Zhuoran Zhuang

Credit assignment in Meta-reinforcement learning (Meta-RL) is still poorly understood. Existing methods either neglect credit assignment to pre-adaptation behavior or implement it naively. This leads to poor sample-efficiency during…

Machine Learning · Computer Science 2022-02-14 Jonas Rothfuss , Dennis Lee , Ignasi Clavera , Tamim Asfour , Pieter Abbeel

To overcome the sparse reward challenge in reinforcement learning (RL) for agents based on large language models (LLMs), we propose Mutual Information Self-Evaluation (MISE), an RL paradigm that utilizes hindsight generative self-evaluation…

Computation and Language · Computer Science 2026-04-14 Jiashu Yao , Heyan Huang , Zeming Liu , Yuhang Guo

Recent self-rewarding large language models (LLM) have successfully applied LLM-as-a-Judge to iteratively improve the alignment performance without the need of human annotations for preference data. These methods commonly utilize the same…

Machine Learning · Computer Science 2025-04-29 Zhaoyang Wang , Weilei He , Zhiyuan Liang , Xuchao Zhang , Chetan Bansal , Ying Wei , Weitong Zhang , Huaxiu Yao

The proliferation of Large Language Models (LLMs) in function calling is pivotal for creating advanced AI agents, yet their large scale hinders widespread adoption, necessitating transferring their capabilities into smaller ones. However,…

Artificial Intelligence · Computer Science 2026-02-25 Jiliang Ni , Jiachen Pu , Zhongyi Yang , Jingfeng Luo , Conggang Hu

Outcome-reward reinforcement learning (RL) is a common and increasingly significant way to refine the step-by-step reasoning of multimodal large language models (MLLMs). In the multiple-choice setting - a dominant format for multimodal…

Computer Vision and Pattern Recognition · Computer Science 2025-11-14 Jiahao Wang , Weiye Xu , Aijun Yang , Wengang Zhou , Lewei Lu , Houqiang Li , Xiaohua Wang , Jinguo Zhu

Audio Large Language Models (ALLMs) are highly vulnerable to real-world noise, which often induces severe semantic drift and hallucinations. Existing robustness methods primarily rely on waveform-level acoustic enhancement, answer-level…

Computation and Language · Computer Science 2026-05-26 Liang Lin , Chunxi Luo , Kaiwen Luo , Jie Zhang , Jin Wang , Yuanhe Zhang , Cai Yuchen , Qiankun Li , Gongli Xi , Zhenhong Zhou , Kun Wang , Junhao Dong

As large language models (LLMs) are deployed in consequential settings such as medical question answering and legal reasoning, the ability to estimate when their outputs are likely to be correct is essential for safe and reliable use,…

Computation and Language · Computer Science 2026-05-22 Fengfei Yu , Ruijia Niu , Dongxia Wu , Yian Ma , Rose Yu

Traditional black-box distillation for Large Vision-Language Models (LVLMs) typically relies on a single teacher response per input, which often yields high-variance responses and format inconsistencies in multimodal or temporal scenarios.…

Computer Vision and Pattern Recognition · Computer Science 2026-03-13 Songlin Li , Xin Zhu , Zechao Guan , Peipeng Chen , Jian Yao

Large Language Models (LLMs) are highly accurate in classification tasks, however, substantial computational and financial costs hinder their large-scale deployment in dynamic environments. Knowledge Distillation (KD) where a LLM "teacher"…

Machine Learning · Computer Science 2025-11-18 Viviana Luccioli , Rithika Iyengar , Ryan Panley , Flora Haberkorn , Xiaoyu Ge , Leland Crane , Nitish Sinha , Seung Jung Lee

Traditional continual learning methods prioritize knowledge retention and focus primarily on mitigating catastrophic forgetting, implicitly assuming that the data distribution of previously learned tasks remains static. This overlooks the…

Machine Learning · Computer Science 2026-02-16 Alif Ashrafee , Jedrzej Kozal , Michal Wozniak , Bartosz Krawczyk

Latent Consistency Distillation (LCD) has emerged as a promising paradigm for efficient text-to-image synthesis. By distilling a latent consistency model (LCM) from a pre-trained teacher latent diffusion model (LDM), LCD facilitates the…

Computer Vision and Pattern Recognition · Computer Science 2024-10-10 Jiachen Li , Weixi Feng , Wenhu Chen , William Yang Wang

Reinforcement learning for multi-step reasoning with large language models (LLMs) typically relies on sparse terminal rewards, which creates a poorly conditioned credit-assignment problem: the final feedback is propagated uniformly across…

Machine Learning · Computer Science 2026-05-26 Fei Ding , Yongkang Zhang , Youwei Wang , Zijian Zeng

Reinforcement learning (RL) for large language model reasoning is frequently hindered by signal loss, a phenomenon where standard uniform sampling with small group sizes fails to uncover informative learning signals for difficult prompts.…

Machine Learning · Computer Science 2025-12-08 Wei Xiong , Chenlu Ye , Baohao Liao , Hanze Dong , Xinxing Xu , Christof Monz , Jiang Bian , Nan Jiang , Tong Zhang

Most deep metric learning (DML) methods employ a strategy that forces all positive samples to be close in the embedding space while keeping them away from negative ones. However, such a strategy ignores the internal relationships of…

Computer Vision and Pattern Recognition · Computer Science 2022-06-20 Zelong Zeng , Fan Yang , Zheng Wang , Shin'ichi Satoh

Improving the multi-step reasoning ability of Large Language Models (LLMs) is a critical yet challenging task. The dominant paradigm, outcome-supervised reinforcement learning (RLVR), rewards only correct final answers, often propagating…

Artificial Intelligence · Computer Science 2025-10-14 Beining Wang , Weihang Su , Hongtao Tian , Tao Yang , Yujia Zhou , Ting Yao , Qingyao Ai , Yiqun Liu

Self-distillation enables language models to learn on-policy from their own trajectories by using the same model as both student and teacher, with the teacher being conditioned on privileged information unavailable to the student. Such…

Empowering large language models (LLMs) with effective tool utilization capabilities is crucial for enabling AI agents to solve complex problems. However, current models face two major limitations: (1) unreliable tool planning and…

Computation and Language · Computer Science 2025-06-06 Zhiyuan Ma , Jiayu Liu , Xianzhen Luo , Zhenya Huang , Qingfu Zhu , Wanxiang Che

Autoregressive models (ARMs) are hindered by slow sequential inference. While masked diffusion models (MDMs) offer a parallel alternative, they suffer from critical drawbacks: high computational overhead from precluding Key-Value (KV)…

Computation and Language · Computer Science 2026-03-06 Jia-Nan Li , Jian Guan , Wei Wu , Chongxuan Li

Self-Distillation Policy Optimization (SDPO) provides dense token-level credit assignment for reinforcement learning with large language models by leveraging the model's own feedback-conditioned predictions as a self-teacher. Unlike GRPO,…

Machine Learning · Computer Science 2026-05-28 Zehao Liu , Yuanpu Cao , Jinghui Chen , Vasant G. Honavar