paper-with-me

Papers

AdaJudge: Adaptive Multi-Perspective Judging for Reward Modeling

2026-01-13 · Yongliang Miao, Yangyang Liang, Mengnan Du arxiv

Reward modeling is essential for aligning large language models with human preferences, yet predominant architectures rely on a static pooling strategy to condense sequences into scalar scores. This paradigm, however, suffers from two key limitations: a static inductive bias that misaligns with task-dependent preference signals, and a representational mismatch, as the backbone's optimization for generation leaves its representations ill-suited to fine-grained discrimination. To address this, we propose AdaJudge, a unified framework that jointly adapts representation and aggregation. AdaJudge first improves backbone representations into a discrimination-oriented space via gated refinement blocks. It then replaces the static readout with an adaptive multi-view pooling module, which dynamically routes and combines evidence. Extensive experiments on RM-Bench and JudgeBench show that AdaJudge outperforms strong off-the-shelf reward models and traditional pooling baselines.

📄 PDF Abstract BibTeX arXiv:2601.08097

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Better Language Model-Based Judging Reward Modeling through Scaling Comprehension Boundaries

2025-08-25 · Meiling Ning, Zhongbao Zhang, Junda Ye, Jiabao Guo 외 arxiv

The emergence of LM-based judging reward modeling, represented by generative reward models, has successfully made reinforcement learning from AI feedback (RLAIF) efficient and scalable. To further advance this paradigm, …

Natural Language UnderstandingNatural Language InferenceReinforcement Learning

SAIL-RL: Guiding MLLMs in When and How to Think via Dual-Reward RL Tuning

2025-11-04 · Fangxun Shu, Yongjie Ye, Yue Liao, Zijian Kang 외 arxiv

We introduce SAIL-RL, a reinforcement learning (RL) post-training framework that enhances the reasoning capabilities of multimodal large language models (MLLMs) by teaching them when and how to think. Existing approaches…

Reinforcement Learning

Judging with Many Minds: Do More Perspectives Mean Less Prejudice?

2025-05-26 · Chiyu Ma, Enpei Zhang, Yilun Zhao, Wenjun Liu 외

LLM-as-Judge has emerged as a scalable alternative to human evaluation, enabling large language models (LLMs) to provide reward signals in trainings. While recent work has explored multi-agent extensions such as multi-ag…

FairJudge: An Adaptive, Debiased, and Consistent LLM-as-a-Judge

2026-02-06 · Bo Yang, Lanfei Feng, Yunkui Chen, Yu Zhang 외 arxiv

Existing LLM-as-a-Judge systems suffer from three fundamental limitations: limited adaptivity to task- and domain-specific evaluation criteria, systematic biases driven by non-semantic cues such as position, length, form…

ARBOR: Online Process Rewards via a Reusable Rubric Buffer for Search Agents

2026-06-02 · Zheng Liu, Longxiang Zhang, Xintong Wang, Zhiang Xu 외 arxiv

LLM-based search agents are trained predominantly with outcome-only reward, leaving the search process itself unsupervised. This signal degenerates on outcome-homogeneous groups where all sampled trajectories share the s…