English
Related papers

Related papers: Self-Induced Outcome Potential: Turn-Level Credit …

200 papers

Through alignment with human preferences, Large Language Models (LLMs) have advanced significantly in generating honest, harmless, and helpful responses. However, collecting high-quality preference data is a resource-intensive and…

Computation and Language · Computer Science 2024-10-10 Qingxiu Dong , Li Dong , Xingxing Zhang , Zhifang Sui , Furu Wei

We study self-rewarding reasoning large language models (LLMs), which can simultaneously generate step-by-step reasoning and evaluate the correctness of their outputs during the inference time-without external feedback. This integrated…

Artificial Intelligence · Computer Science 2025-02-28 Wei Xiong , Hanning Zhang , Chenlu Ye , Lichang Chen , Nan Jiang , Tong Zhang

In this paper, we propose a novel approach where agents can form swarms to produce high-quality responses effectively. This is accomplished by utilizing agents capable of data inference and ranking, which can be effectively implemented…

Multiagent Systems · Computer Science 2024-09-16 Vladyslav Larin , Ivan Nikitin , Alexander Firsov

Reinforcement learning (RL) holds significant promise for training LLM agents to handle complex, goal-oriented tasks that require multi-step interactions with external environments. However, a critical challenge when applying RL to these…

Computation and Language · Computer Science 2025-05-28 Hanlin Wang , Chak Tou Leong , Jiashuo Wang , Jian Wang , Wenjie Li

Deploying GRPO on Flow Matching models has proven effective for text-to-image generation. However, existing paradigms typically propagate an outcome-based reward to all preceding denoising steps without distinguishing the local effect of…

Computer Vision and Pattern Recognition · Computer Science 2026-02-09 Yunze Tong , Mushui Liu , Canyu Zhao , Wanggui He , Shiyi Zhang , Hongwei Zhang , Peng Zhang , Jinlong Liu , Ju Huang , Jiamang Wang , Hao Jiang , Pipei Huang

We present Future-KL Influenced Policy Optimization (FIPO), a reinforcement learning algorithm designed to overcome reasoning bottlenecks in large language models. While GRPO style training scales effectively, it typically relies on…

Machine Learning · Computer Science 2026-04-01 Chiyu Ma , Shuo Yang , Kexin Huang , Jinda Lu , Haoming Meng , Shangshang Wang , Bolin Ding , Soroush Vosoughi , Guoyin Wang , Jingren Zhou

Large Language Models (LLMs) can achieve inflated scores on multiple-choice tasks by exploiting inherent biases in option positions or labels, rather than demonstrating genuine understanding. This study introduces SCOPE, an evaluation…

Computation and Language · Computer Science 2025-08-05 Wonjun Jeong , Dongseok Kim , Taegkeun Whangbo

Pre-trained large language models (LLMs) can be tailored to adhere to human instructions through instruction tuning. However, due to shifts in the distribution of test-time data, they may not always execute instructions accurately,…

Computation and Language · Computer Science 2024-09-04 Hai Ye , Hwee Tou Ng

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

The next-token prediction (NTP) objective trains language models to predict a single token at each step, even though many continuations can express the same meaning. For example, in the sentence ``this sticker can be placed here'',…

Computation and Language · Computer Science 2026-05-26 Christine Zhang , Dan Jurafsky , Chen Shani

A fundamental challenge in formal theorem proving by LLMs is the lack of high-quality training data. Although reinforcement learning or expert iteration partially mitigates this issue by alternating between LLM generating proofs and…

Machine Learning · Computer Science 2025-03-24 Kefan Dong , Tengyu Ma

The opaqueness of deep NLP models has motivated the development of methods for interpreting how deep models predict. Recently, work has introduced hierarchical attribution, which produces a hierarchical clustering of words, along with an…

Computation and Language · Computer Science 2022-10-25 Yiming Ju , Yuanzhe Zhang , Kang Liu , Jun Zhao

Recent advances in Large Language Model (LLM) agents have demonstrated their promising general capabilities. However, their performance in specialized real-world domains often degrades due to challenges in effectively integrating external…

Computation and Language · Computer Science 2025-10-10 Yuzheng Cai , Siqi Cai , Yuchen Shi , Zihan Xu , Lichao Chen , Yulei Qin , Xiaoyu Tan , Gang Li , Zongyi Li , Haojia Lin , Yong Mao , Ke Li , Xing Sun

We present an outcome-driven fine-tuning framework that enhances the forecasting capabilities of large language models (LLMs) without relying on human-curated reasoning samples. Our method leverages model self-play to generate pairs of…

Computation and Language · Computer Science 2025-02-11 Benjamin Turtel , Danny Franklin , Philipp Schoenegger

Current techniques for post-training Large Language Models (LLMs) rely either on costly human supervision or on external verifiers to boost performance on tasks such as mathematical reasoning and code generation. However, as LLMs improve…

Computation and Language · Computer Science 2026-01-21 Mukesh Ghimire , Aosong Feng , Liwen You , Youzhi Luo , Fang Liu , Xuan Zhu

The paper compares two generic techniques for deriving lower bounds and impossibility results in distributed computing. First, we prove a speedup theorem (a-la Brandt, 2019), for wait-free colorless algorithms, aiming at capturing the…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-08-09 Hagit Attiya , Pierre Fraigniaud , Ami Paz , Sergio Rajsbaum

Inference-time alignment techniques offer a lightweight alternative or complement to costly reinforcement learning, while enabling continual adaptation as alignment objectives and reward targets evolve. Existing theoretical analyses justify…

Machine Learning · Computer Science 2026-05-14 Ye Wang , Jing Liu , Toshiaki Koike-Akino

In this paper, we study Interaction-Grounded Learning (IGL) [Xie et al., 2021], a paradigm designed for realistic scenarios where the learner receives indirect feedback generated by an unknown mechanism, rather than explicit numerical…

Machine Learning · Computer Science 2026-02-10 Mengxiao Zhang , Yuheng Zhang , Haipeng Luo , Paul Mineiro

Training task-oriented dialogue systems typically requires turn-level annotations for interacting with their APIs: e.g. a dialogue state and the system actions taken at each step. These annotations can be costly to produce, error-prone, and…

Computation and Language · Computer Science 2024-10-17 Brendan King , Jeffrey Flanigan

Practical mechanisms often limit agent reports to constrained formats like trades or orderings, potentially limiting the information agents can express. We propose a novel class of mechanisms that elicit agent reports in natural language…

Computer Science and Game Theory · Computer Science 2024-07-11 Nicolas Della Penna