English
Related papers

Related papers: When Rubrics Fail: Error Enumeration as Reward in …

200 papers

We introduce the "Incremental Implicitly-Refined Classi-fication (IIRC)" setup, an extension to the class incremental learning setup where the incoming batches of classes have two granularity levels. i.e., each sample could have a…

Computer Vision and Pattern Recognition · Computer Science 2021-01-13 Mohamed Abdelsalam , Mojtaba Faramarzi , Shagun Sodhani , Sarath Chandar

Recent advances in reinforcement learning (RL) have significantly improved the complex reasoning capabilities of large language models (LLMs). Despite these successes, existing methods mainly focus on single-domain RL (e.g., mathematics)…

Artificial Intelligence · Computer Science 2025-11-20 Baolong Bi , Shenghua Liu , Yiwei Wang , Siqian Tong , Lingrui Mei , Yuyao Ge , Yilong Xu , Jiafeng Guo , Xueqi Cheng

Rubric-based evaluation is widely used in LLM benchmarks and training pipelines for open-ended, less verifiable tasks. While prior work has demonstrated the effectiveness of rubrics using downstream signals such as reinforcement learning…

Artificial Intelligence · Computer Science 2026-04-22 Zhengyang Qi , Charles Dickens , Derek Pham , Amanda Dsouza , Armin Parchami , Frederic Sala , Paroma Varma

Reinforcement learning with verifiable rewards (RLVR) has become a powerful paradigm for improving language models on reasoning-intensive tasks, but its effectiveness is often limited by exploration. For example, models often fail on hard…

Machine Learning · Computer Science 2026-05-26 Ruitao Liu , Qinghao Hu , Alex Hu , Yecheng Wu , Shang Yang , Luke J. Huang , Zhuoyang Zhang , Han Cai , Song Han

Recent reinforcement learning (RL) approaches have advanced radiology report generation (RRG), yet two core limitations persist: (1) report-level rewards offer limited evidence-grounded guidance for clinical faithfulness; and (2) current…

Machine Learning · Computer Science 2026-04-16 Qin Zhou , Guoyan Liang , Qianyi Yang , Jingyuan Chen , Sai Wu , Chang Yao , Zhe Wang

Data entry systems remain structurally vulnerable to categorical misclassifications, particularly in small and medium sized enterprises (SMEs). When nominal categories exhibit semantic or morphological proximity, human machine interaction…

Computation and Language · Computer Science 2026-05-13 Ricardo Raúl Palma , Mauro Anibal Benetti , Fabricio Orlando Sanchez Varretti

Reinforcement learning with verifiable rewards (RLVR) succeeds in reasoning tasks (e.g., math and code) by checking the final verifiable answer (i.e., a verifiable dot signal). However, extending this paradigm to open-ended generation is…

Computation and Language · Computer Science 2026-01-27 Yuxin Jiang , Yufei Wang , Qiyuan Zhang , Xingshan Zeng , Liangyou Li , Jierun Chen , Chaofan Tao , Haoli Bai , Lifeng Shang

Reinforcement Learning with Verifiable Rewards (RLVR) has driven substantial progress in reasoning-intensive domains like mathematics. However, optimizing open-ended generation remains challenging due to the lack of ground truth. While…

Artificial Intelligence · Computer Science 2026-01-29 Sunzhu Li , Jiale Zhao , Miteto Wei , Huimin Ren , Yang Zhou , Jingwen Yang , Shunyu Liu , Kaike Zhang , Wei Chen

Most reward models for visual generation reduce rich human judgments to a single unexplained score, discarding the reasoning that underlies preference. We show that teaching reward models to produce explicit, multi-dimensional critiques…

Artificial Intelligence · Computer Science 2026-04-15 Haozhe Wang , Cong Wei , Weiming Ren , Jiaming Liu , Fangzhen Lin , Wenhu Chen

Language models trained with reinforcement learning (RL) can engage in reward hacking--the exploitation of unintended strategies for high reward--without revealing this behavior in their chain-of-thought reasoning. This makes the detection…

Computation and Language · Computer Science 2025-07-15 Miles Turpin , Andy Arditi , Marvin Li , Joe Benton , Julian Michael

Open-ended evaluation is essential for deploying large language models in real-world settings. In studying HealthBench, we observe that using the model itself as a grader and generating rubric-based reward signals substantially improves…

Computation and Language · Computer Science 2025-10-01 Zhiling Ye , Yun Yue , Haowen Wang , Xudong Han , Jiadi Jiang , Cheng Wei , Lei Fan , Jiaxin Liang , Shuowen Zhang , Ji Li , Chunxiao Guo , Jian Wang , Peng Wei , Jinjie Gu

While in-context learning with large language models (LLMs) has shown impressive performance, we have discovered a unique miscalibration behavior where both correct and incorrect predictions are assigned the same level of confidence. We…

Computation and Language · Computer Science 2024-10-04 Wei Cheng , Tianlu Wang , Yanmin Ji , Fan Yang , Keren Tan , Yiyu Zheng

Reinforcement Learning with Verifiable Rewards (RLVR) replaces costly human labeling with automated verifiers. To reduce verifier hacking, many RLVR systems binarize rewards to $\{0,1\}$, but imperfect verifiers inevitably introduce…

Machine Learning · Computer Science 2026-05-25 Xin-Qiang Cai , Wei Wang , Feng Liu , Tongliang Liu , Gang Niu , Masashi Sugiyama

Most reinforcement learning (RL) methods for training large language models (LLMs) require ground-truth labels or task-specific verifiers, limiting scalability when correctness is ambiguous or expensive to obtain. We introduce Reinforcement…

Neural and Evolutionary Computing · Computer Science 2026-01-30 Micah Rentschler , Jesse Roberts

Vision-Language Models (VLMs) frequently suffer from visual perception errors and hallucinations that compromise answer accuracy in complex reasoning tasks. Reinforcement Learning with Verifiable Rewards (RLVR) offers a promising solution…

Computer Vision and Pattern Recognition · Computer Science 2026-05-05 Yin Zhang , Jiaxuan Zhao , Zonghan Wu , Zengxiang Li , Junfeng Fang , Kun Wang , Qingsong Wen , Yilei Shao

Recent advances such as DeepSeek R1-Zero highlight the effectiveness of incentive training, a reinforcement learning paradigm that computes rewards solely based on the final answer part of a language model's output, thereby encouraging the…

Computation and Language · Computer Science 2025-09-04 Wei Liu , Siya Qi , Xinyu Wang , Chen Qian , Yali Du , Yulan He

Language models often generate factually incorrect information unsupported by their training data, a phenomenon known as extrinsic hallucination. Existing mitigation approaches often degrade performance on open-ended generation and…

Computation and Language · Computer Science 2025-10-21 Tong Chen , Akari Asai , Luke Zettlemoyer , Hannaneh Hajishirzi , Faeze Brahman

Different from its counterpart outcome reward models (ORMs), which evaluate the entire responses, a process reward model (PRM) scores a reasoning trajectory step by step, providing denser and more fine grained rewards. However, training a…

Machine Learning · Computer Science 2024-12-04 Lifan Yuan , Wendi Li , Huayu Chen , Ganqu Cui , Ning Ding , Kaiyan Zhang , Bowen Zhou , Zhiyuan Liu , Hao Peng

As reinforcement Learning with Verifiable Rewards (RLVR) has become the dominant paradigm for scaling reasoning capabilities in LLMs, a new failure mode emerges: LLMs gaming verifiers. We study this phenomenon on inductive reasoning tasks,…

Recent advancements in the field of natural language generation have facilitated the use of large language models to assess the quality of generated text. Although these models have shown promising results in tasks such as machine…

Artificial Intelligence · Computer Science 2024-01-23 Terry Yue Zhuo
‹ Prev 1 4 5 6 7 8 10 Next ›