pith. sign in

arxiv: 2605.30244 · v1 · pith:GSHUNBCTnew · submitted 2026-05-28 · 💻 cs.CV · cs.AI

Reinforcement Learning with Robust Rubric Rewards

Ya-Qi Yu , Hao Wang , Fangyu Hong , Xiangyang Qu , Gaojie Wu , Qiaoyu Luo , Nuo Xu , Huixin Wang
show 10 more authors
Wuheng Xu Yongxin Liao Zihao Chen Haonan Li Ziming Li Dezhi Peng Minghui Liao Jihao Wu Haoyu Ren Dandan Tu
This is my paper

Pith reviewed 2026-06-29 07:37 UTC · model grok-4.3

classification 💻 cs.CV cs.AI
keywords reinforcement learningvision-language modelsrubric rewardsverifiable rewardsmulti-criteria supervisionreward hacking mitigationRLVR extension
0
0 comments X

The pith

Reinforcement learning with robust rubric rewards extends verifiable supervision from whole tasks to individual criteria in vision-language models.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces RLR³ to address the limitation that standard RLVR works only for fully deterministically checkable tasks while many vision-language problems require fine-grained multi-criteria feedback on perception, reasoning, and constraints. It routes instance-specific rubrics through an LLM extractor plus deterministic verifier for checkable criteria or an LLM judge for the rest, applying minimal exposure masking so ground truths stay hidden from extractors and images stay hidden from judges. Hierarchical aggregation then weights essential criteria more heavily, and score saturation within rollout batches is mitigated. On the Qwen3-VL-30B-A3B model across 15 benchmarks this produces a 4.7-point gain over the base model and beats both RLVR and the official instruct-to-thinking variant while controlled audits show reduced false-positive exploitation.

Core claim

RLR³ extends RLVR from task-level verification to criterion-level verification by executing instance-specific rubrics along two protected paths—an LLM-as-extractor paired with a deterministic verifier, or an LLM-as-Judge—while enforcing minimal exposure masking of ground truths and images. Hierarchical aggregation prioritizes essential criteria and saturation mitigation prevents score collapse within rollout groups. The resulting rewards remain faithful enough that the method delivers consistent gains over RLVR baselines and the official model gap on the evaluated vision-language benchmarks.

What carries the argument

Dual rubric execution paths (LLM extractor plus deterministic verifier, or LLM judge) with minimal exposure masking, which together produce criterion-level scores that support online RL without introducing exploitable false positives.

If this is right

  • RLVR can be extended from binary task success to graded multi-criteria supervision without losing deterministic guarantees on verifiable parts.
  • Minimal exposure masking plus hierarchical aggregation together reduce score saturation and reward hacking in rubric-based training.
  • The same rubric infrastructure yields measurable gains on 15 diverse vision-language benchmarks for a 30B-scale model.
  • Controlled audits become a practical way to certify that the added LLM components do not increase false-positive exploitation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The dual-path design may transfer to other partially verifiable domains such as code generation or mathematical proof steps where some sub-criteria admit deterministic checks.
  • Minimal exposure could be tested as a general safeguard in any LLM-mediated reward pipeline to limit information leakage to the policy.
  • If the audits hold, rubric rewards might become a scalable middle ground between pure outcome supervision and expensive human preference data.

Load-bearing premise

That routing rubrics through the extractor-verifier or judge paths plus minimal exposure masking produces faithful criterion scores that do not create new exploitable false positives beyond those checked in the audits.

What would settle it

A controlled audit or new benchmark run in which the policy achieves high rubric scores by exploiting patterns visible to the extractor or judge yet fails to satisfy the stated criteria when the ground truth is revealed.

Figures

Figures reproduced from arXiv: 2605.30244 by Dandan Tu, Dezhi Peng, Fangyu Hong, Gaojie Wu, Haonan Li, Hao Wang, Haoyu Ren, Huixin Wang, Jihao Wu, Minghui Liao, Nuo Xu, Qiaoyu Luo, Wuheng Xu, Xiangyang Qu, Ya-Qi Yu, Yongxin Liao, Zihao Chen, Ziming Li.

Figure 1
Figure 1. Figure 1: Overview of RLR3 . Instance-specific rubrics turn response quality assessment into concrete criteria and criterion-level rewards. Verifiable criteria are routed to a text-only LLM-as-an-extractor followed by deterministic verification, while fuzzy criteria are routed to a text-only LLM-as-a-Judge. We propose Reinforcement Learning with Robust Rubric Rewards (RLR3 ), a framework that extends RLVR from task-… view at source ↗
Figure 2
Figure 2. Figure 2: Training trajectories across benchmarks. Each subplot reports benchmark performance over training checkpoints for RLVR and RLR3 , with the dashed gray line showing the base model [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The diagram illustrates the components of a coal-fired power plant. [PITH_FULL_IMAGE:figures/full_fig_p020_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The table lists books and their prices. Question & Ground Truth Q: Which book is the least expensive? A: book about Asia Model Response (Symbolic Bluff) To determine the least expensive book, we must construct a cost minimization framework over the discrete set of available literature B = {b1, b2, . . . , bn}. Let the price function P(bi) be defined for each book bi in the catalog. We initialize the optimi… view at source ↗
read the original abstract

While Reinforcement Learning with Verifiable Rewards (RLVR) is effective for deterministically checkable tasks, many vision-language tasks are partially verifiable, demanding multi-criteria supervision (e.g., perceptual details, reasoning steps, and constraints). Rubrics provide a natural interface for this fine-grained supervision, but their effectiveness depends on the execution accuracy during online RL. We propose Reinforcement Learning with Robust Rubric Rewards ($\text{RLR}^3$), extending RLVR from task-level verification to criterion-level verification. $\text{RLR}^3$ routes instance-specific rubrics through two execution paths: an LLM-as-an-extractor paired with a deterministic verifier, or an LLM-as-a-Judge for non-verifiable criteria. To ensure faithful scoring, $\text{RLR}^3$ introduce a minimal exposure strategy that masks ground truths from extractors and images from judges. Furthermore, $\text{RLR}^3$ employs hierarchical aggregation to prioritize essential criteria over additional criteria, and mitigates score saturation within rollout groups. Evaluated on Qwen3-VL-30B-A3B across 15 benchmarks, $\text{RLR}^3$ consistently outperforms RLVR, yielding a 4.7-point improvement over the base model and exceeding the official instruct-to-thinking model gap. Controlled audits confirm our deterministic verification and minimal exposure significantly reduce exploitable false positives.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper proposes Reinforcement Learning with Robust Rubric Rewards (RLR³) as an extension of RLVR for vision-language tasks. It routes instance-specific rubrics through an LLM-as-extractor with deterministic verifier or LLM-as-Judge paths, applies minimal exposure (masking ground truth from extractors and images from judges), uses hierarchical aggregation to prioritize criteria, and mitigates score saturation. The central claim is that RLR³ consistently outperforms RLVR on Qwen3-VL-30B-A3B across 15 benchmarks, delivering a 4.7-point average gain over the base model that exceeds the official instruct-to-thinking gap, with controlled audits showing reduced exploitable false positives.

Significance. If the empirical gains and audit results prove robust, the approach offers a practical route to criterion-level rewards for partially verifiable VL tasks, potentially improving upon task-level RLVR by reducing reward hacking while preserving fine-grained supervision. The minimal-exposure masking and hierarchical aggregation are concrete engineering contributions that could generalize beyond the evaluated model.

major comments (2)
  1. [Abstract] Abstract: The headline result of a 4.7-point improvement (and outperformance of the instruct-to-thinking gap) is reported without error bars, per-benchmark tables, or any statistical significance tests. This is load-bearing for the claim of consistent outperformance across 15 benchmarks.
  2. [Abstract] Abstract: The assertion that 'controlled audits confirm our deterministic verification and minimal exposure significantly reduce exploitable false positives' provides no quantitative details on audit scale, rubric diversity, or measured judge accuracy under image masking for perceptual or constraint criteria. This directly affects the central assumption that the routing paths produce faithful scores without introducing new leakage.
minor comments (1)
  1. [Abstract] Grammatical error in the abstract: 'RLR³ introduce a minimal exposure strategy' should read 'RLR³ introduces a minimal exposure strategy'.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and agree to revisions that improve the transparency of the reported results and audit details.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The headline result of a 4.7-point improvement (and outperformance of the instruct-to-thinking gap) is reported without error bars, per-benchmark tables, or any statistical significance tests. This is load-bearing for the claim of consistent outperformance across 15 benchmarks.

    Authors: We agree that the abstract would benefit from explicit pointers to the supporting analyses. The full manuscript reports per-benchmark scores across all 15 benchmarks in Table 2 (showing gains on every benchmark) along with the average improvement. Standard deviations from multiple random seeds appear in the appendix, and paired statistical significance tests are summarized in Appendix B. We will revise the abstract to reference these elements directly after the headline claim. revision: yes

  2. Referee: [Abstract] Abstract: The assertion that 'controlled audits confirm our deterministic verification and minimal exposure significantly reduce exploitable false positives' provides no quantitative details on audit scale, rubric diversity, or measured judge accuracy under image masking for perceptual or constraint criteria. This directly affects the central assumption that the routing paths produce faithful scores without introducing new leakage.

    Authors: The manuscript already contains the requested quantitative details in Section 4.3 (audit scale, rubric diversity across criteria types, and judge accuracy under image masking). To make this evidence visible at the abstract level, we will revise the relevant sentence to incorporate a concise summary of those quantitative findings. revision: yes

Circularity Check

0 steps flagged

No circularity: procedural method with external benchmark results

full rationale

The paper describes a procedural extension of RLVR to rubric-based criterion verification using LLM extractors, deterministic verifiers, LLM judges, minimal exposure masking, and hierarchical aggregation. No equations, derivations, or fitted parameters appear in the provided text. Performance claims rest on direct evaluations across 15 benchmarks on Qwen3-VL-30B-A3B, which are independent external measurements rather than quantities defined by the method itself. No self-citations, ansatzes, or uniqueness theorems are invoked in a load-bearing manner. The derivation chain is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The method relies on the unstated premise that LLM extractors and judges can be made sufficiently reliable via masking; no free parameters are named in the abstract, but hierarchical aggregation implies at least one weighting choice between essential and additional criteria.

free parameters (1)
  • criterion priority weights
    Hierarchical aggregation requires choosing relative importance between essential and additional criteria; value not stated in abstract.
axioms (1)
  • domain assumption LLM-as-extractor and LLM-as-Judge produce sufficiently accurate criterion scores when ground truth and images are masked
    Invoked to justify the two execution paths and minimal exposure strategy.

pith-pipeline@v0.9.1-grok · 5832 in / 1458 out tokens · 23009 ms · 2026-06-29T07:37:20.988448+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references · 29 canonical work pages · 14 internal anchors

  1. [1]

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, Bei Feng, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chong Ruan, Damai Dai, Deli Chen, Dongjie Ji, ...

    2025

  2. [2]

    Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Mingchuan Zhang, Y . K. Li, Y . Wu, and Daya Guo. Deepseekmath: Pushing the limits of mathematical reasoning in open language models.CoRR, abs/2402.03300, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  3. [3]

    Kimi K2: Open Agentic Intelligence

    Kimi Team. Kimi K2: open agentic intelligence.CoRR, abs/2507.20534, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  4. [4]

    GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models

    GLM. GLM-4.5: agentic, reasoning, and coding (ARC) foundation models.CoRR, abs/2508.06471, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  5. [5]

    VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model

    Haozhan Shen, Peng Liu, Jingcheng Li, Chunxin Fang, Yibo Ma, Jiajia Liao, Qiaoli Shen, Zilun Zhang, Kangjia Zhao, Qianqian Zhang, Ruochen Xu, and Tiancheng Zhao. VLM-R1: A stable and generalizable r1-style large vision-language model.CoRR, abs/2504.07615, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  6. [6]

    VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning

    Haozhe Wang, Chao Qu, Zuming Huang, Wei Chu, Fangzhen Lin, and Wenhu Chen. Vl- rethinker: Incentivizing self-reflection of vision-language models with reinforcement learning. CoRR, abs/2504.08837, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  7. [7]

    Kimi K2.5: Visual Agentic Intelligence

    Kimi Team. Kimi K2.5: visual agentic intelligence.CoRR, abs/2602.02276, 2026

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  8. [8]

    GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

    Wenyi Hong, Wenmeng Yu, Xiaotao Gu, Guo Wang, Guobing Gan, Haomiao Tang, Jiale Cheng, Ji Qi, Junhui Ji, Lihang Pan, Shuaiqi Duan, Weihan Wang, Yan Wang, Yean Cheng, Zehai He, Zhe Su, Zhen Yang, Ziyang Pan, Aohan Zeng, Baoxu Wang, Boyan Shi, Changyu Pang, Chenhui Zhang, Da Yin, Fan Yang, Guoqing Chen, Jiazheng Xu, Jiali Chen, Jing Chen, Jinhao Chen, Jingha...

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  9. [9]

    Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains

    Anisha Gunjal, Anthony Wang, Elaine Lau, Vaskar Nath, Bing Liu, and Sean Hendryx. Rubrics as rewards: Reinforcement learning beyond verifiable domains.CoRR, abs/2507.17746, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  10. [10]

    Checklists are better than reward models for aligning language models.CoRR, abs/2507.18624, 2025

    Vijay Viswanathan, Yanchao Sun, Shuang Ma, Xiang Kong, Meng Cao, Graham Neubig, and Tongshuang Wu. Checklists are better than reward models for aligning language models.CoRR, abs/2507.18624, 2025

    work page arXiv 2025

  11. [11]

    Preprint, arXiv:2508.12790

    Zenan Huang, Yihong Zhuang, Guoshan Lu, Zeyu Qin, Haokai Xu, Tianyu Zhao, Ru Peng, Jiaqi Hu, Zhanming Shen, Xiaomeng Hu, Xijun Gu, Peiyi Tu, Jiaxin Liu, Wenyu Chen, Yuzhuo Fu, Zhiting Fan, Yanmei Gu, Yuanyuan Wang, Zhengkai Yang, Jianguo Li, and Junbo Zhao. Reinforcement learning with rubric anchors.CoRR, abs/2508.12790, 2025

    work page arXiv 2025

  12. [12]

    Visual Preference Optimization with Rubric Rewards

    Ya-Qi Yu, Fangyu Hong, Xiangyang Qu, Hao Wang, Gaojie Wu, Qiaoyu Luo, Nuo Xu, Huixin Wang, Wuheng Xu, Yongxin Liao, Zihao Chen, Haonan Li, Ziming Li, Dezhi Peng, Minghui Liao, Jihao Wu, Haoyu Ren, and Dandan Tu. Visual preference optimization with rubric rewards. CoRR, abs/2604.13029, 2026

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  13. [13]

    Shu Pu, Yaochen Wang, Dongping Chen, Yuhang Chen, Guohao Wang, Qi Qin, Zhongyi Zhang, Zhiyuan Zhang, Zetong Zhou, Shuang Gong, Yi Gui, Yao Wan, and Philip S. Yu. Judge anything: MLLM as a judge across any modality. In Luiza Antonie, Jian Pei, Xiaohui Yu, Flavio Chierichetti, Hady W. Lauw, Yizhou Sun, and Srinivasan Parthasarathy, editors,Proceedings of th...

    2025

  14. [14]

    Multi-Crit: Benchmarking multimodal judges on pluralistic criteria-following.arXiv preprint arXiv:2511.21662, 2025

    Tianyi Xiong, Yi Ge, Ming Li, Zuolong Zhang, Pranav Kulkarni, Kaishen Wang, Qi He, Zeying Zhu, Chenxi Liu, Ruibo Chen, Tong Zheng, Yanshuo Chen, Xiyao Wang, Renrui Zhang, Wenhu Chen, and Heng Huang. Multi-crit: Benchmarking multimodal judges on pluralistic criteria-following.CoRR, abs/2511.21662, 2025

    work page arXiv 2025

  15. [15]

    Llm- rubric: A multidimensional, calibrated approach to automated evaluation of natural language texts

    Helia Hashemi, Jason Eisner, Corby Rosset, Benjamin Van Durme, and Chris Kedzie. Llm- rubric: A multidimensional, calibrated approach to automated evaluation of natural language texts. In Lun-Wei Ku, Andre Martins, and Vivek Srikumar, editors,Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), A...

    2024

  16. [16]

    Paperbench: Evaluating ai’s ability to replicate AI research

    Giulio Starace, Oliver Jaffe, Dane Sherburn, James Aung, Jun Shern Chan, Leon Maksin, Rachel Dias, Evan Mays, Benjamin Kinsella, Wyatt Thompson, Johannes Heidecke, Amelia Glaese, and Tejal Patwardhan. Paperbench: Evaluating ai’s ability to replicate AI research. In Aarti Singh, Maryam Fazel, Daniel Hsu, Simon Lacoste-Julien, Felix Berkenkamp, Tegan Mahara...

    2025

  17. [17]

    Rahul K. Arora, Jason Wei, Rebecca Soskin Hicks, Preston Bowman, Joaquin Quiñonero Candela, Foivos Tsimpourlas, Michael Sharman, Meghan Shah, Andrea Vallone, Alex Beutel, Johannes Heidecke, and Karan Singhal. Healthbench: Evaluating large language models towards improved human health.CoRR, abs/2505.08775, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  18. [18]

    ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge

    Zhilin Wang, Jaehun Jung, Ximing Lu, Shizhe Diao, Ellie Evans, Jiaqi Zeng, Pavlo Molchanov, Yejin Choi, Jan Kautz, and Yi Dong. Profbench: Multi-domain rubrics requiring professional knowledge to answer and judge.CoRR, abs/2510.18941, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  19. [19]

    Prbench: Large-scale expert rubrics for evaluating high-stakes professional reasoning.CoRR, abs/2511.11562, 2025

    Afra Feyza Akyürek, Advait Gosai, Chen Bo Calvin Zhang, Vipul Gupta, Jaehwan Jeong, Anisha Gunjal, Tahseen Rabbani, Maria Mazzone, David Randolph, Mohammad Mahmoudi Meymand, Gurshaan Chattha, Paula Rodriguez, Diego Mares, Pavit Singh, Michael Liu, Subodh Chawla, Pete Cline, Lucy Ogaz, Ernesto Hernandez, Zihao Wang, Pavi Bhatter, Marcos Ayestaran, Bing Liu...

    work page arXiv 2025

  20. [20]

    Breaking the exploration bottleneck: Rubric-scaffolded reinforcement learning for general llm reasoning.arXiv preprint arXiv:2508.16949,

    Yang Zhou, Sunzhu Li, Shunyu Liu, Wenkai Fang, Jiale Zhao, Jingwen Yang, Jianwei Lv, Kongcheng Zhang, Yihe Zhou, Hengtong Lu, Wei Chen, Yan Xie, and Mingli Song. Breaking the exploration bottleneck: Rubric-scaffolded reinforcement learning for general LLM reasoning. CoRR, abs/2508.16949, 2025

    work page arXiv 2025

  21. [21]

    Alternating reinforcement learning for rubric-based reward modeling in non-verifiable llm post-training.arXiv preprint arXiv:2602.01511,

    Ran Xu, Tianci Liu, Zihan Dong, Tony Yu, Ilgee Hong, Carl Yang, Linjun Zhang, Tao Zhao, and Haoyu Wang. Alternating reinforcement learning for rubric-based reward modeling in non-verifiable llm post-training.CoRR, abs/2602.01511, 2026

    work page arXiv 2026

  22. [22]

    InInternational Con- ference on Learning Representations, volume 2024, pages 29927–29962

    Tianci Liu, Ran Xu, Tony Yu, Ilgee Hong, Carl Yang, Tuo Zhao, and Haoyu Wang. Openrubrics: Towards scalable synthetic rubric generation for reward modeling and LLM alignment.CoRR, abs/2510.07743, 2025

    work page arXiv 2025

  23. [23]

    Online rubrics elicitation from pairwise comparisons.arXiv preprint arXiv:2510.07284,

    MohammadHossein Rezaei, Robert Vacareanu, Zihao Wang, Clinton Wang, Bing Liu, Yunzhong He, and Afra Feyza Akyürek. Online rubrics elicitation from pairwise comparisons.CoRR, abs/2510.07284, 2025

    work page arXiv 2025

  24. [24]

    Auto-rubric: Learning from implicit weights to explicit rubrics for reward modeling.arXiv preprint arXiv:2510.17314,

    Lipeng Xie, Sen Huang, Zhuo Zhang, Anni Zou, Yunpeng Zhai, Dingchao Ren, Kezun Zhang, Haoyuan Hu, Boyin Liu, Haoran Chen, Zhaoyang Liu, and Bolin Ding. Auto-rubric: Learning to extract generalizable criteria for reward modeling.CoRR, abs/2510.17314, 2025

    work page arXiv 2025

  25. [25]

    Rubrichub: A comprehensive and highly discriminative rubric dataset via automated coarse-to-fine generation.CoRR, abs/2601.08430, 2026

    Sunzhu Li, Jiale Zhao, Miteto Wei, Huimin Ren, Yang Zhou, Jingwen Yang, Shunyu Liu, Kaike Zhang, and Wei Chen. Rubrichub: A comprehensive and highly discriminative rubric dataset via automated coarse-to-fine generation.CoRR, abs/2601.08430, 2026

    work page arXiv 2026

  26. [26]

    Omni-rrm: Advancing omni reward modeling via automatic rubric-grounded preference synthe- sis.arXiv preprint arXiv:2602.00846, 2026

    Zicheng Kong, Dehua Ma, Zhenbo Xu, Alven Yang, Yiwei Ru, Haoran Wang, Zixuan Zhou, Fuqing Bie, Liuyu Xiang, Huijia Wu, Jian Zhao, and Zhaofeng He. Omni-rrm: Advancing omni reward modeling via automatic rubric-grounded preference synthesis.CoRR, abs/2602.00846, 2026

    work page arXiv 2026

  27. [27]

    DAPO: An Open-Source LLM Reinforcement Learning System at Scale

    Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Tiantian Fan, Gaohong Liu, Lingjun Liu, Xin Liu, Haibin Lin, Zhiqi Lin, Bole Ma, Guangming Sheng, Yuxuan Tong, Chi Zhang, Mofan Zhang, Wang Zhang, Hang Zhu, Jinhua Zhu, Jiaze Chen, Jiangjie Chen, Chengyi Wang, Hongli Yu, Weinan Dai, Yuxuan Song, Xiangpeng Wei, Hao Zhou, Jingjing Liu, W...

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  28. [28]

    Openmmreasoner: Pushing the frontiers for multimodal reasoning with an open and general recipe.CoRR, abs/2511.16334, 2025

    Kaichen Zhang, Keming Wu, Zuhao Yang, Bo Li, Kairui Hu, Bin Wang, Ziwei Liu, Xingxuan Li, and Lidong Bing. Openmmreasoner: Pushing the frontiers for multimodal reasoning with an open and general recipe.CoRR, abs/2511.16334, 2025

    work page arXiv 2025

  29. [29]

    Deepvision-103k: A visually diverse, broad-coverage, and verifiable mathematical dataset for multimodal reasoning.CoRR, abs/2602.16742, 2026

    Haoxiang Sun, Lizhen Xu, Bing Zhao, Wotao Yin, Wei Wang, Boyu Yang, Rui Wang, and Hu Wei. Deepvision-103k: A visually diverse, broad-coverage, and verifiable mathematical dataset for multimodal reasoning.CoRR, abs/2602.16742, 2026

    work page arXiv 2026

  30. [30]

    Runqi Qiao, Qiuna Tan, Guanting Dong, Minhui Wu, Chong Sun, Xiaoshuai Song, Jiapeng Wang, Zhuoma Gongque, Shanglin Lei, Yifan Zhang, Zhe Wei, Miaoxuan Zhang, Runfeng Qiao, Xiao Zong, Yida Xu, Peiqing Yang, Zhimin Bao, Muxi Diao, Chen Li, and Honggang Zhang. We-math: Does your large multimodal model achieve human-like mathematical reasoning? In Wanxiang Ch...

    2025

  31. [31]

    Dynamath: A dynamic visual benchmark for evaluating mathematical reasoning robustness of vision language models

    Chengke Zou, Xingang Guo, Rui Yang, Junyu Zhang, Bin Hu, and Huan Zhang. Dynamath: A dynamic visual benchmark for evaluating mathematical reasoning robustness of vision language models. InThe Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net, 2025

    2025

  32. [32]

    Measuring multimodal mathematical reasoning with math-vision dataset

    Ke Wang, Junting Pan, Weikang Shi, Zimu Lu, Houxing Ren, Aojun Zhou, Mingjie Zhan, and Hongsheng Li. Measuring multimodal mathematical reasoning with math-vision dataset. In Amir Globersons, Lester Mackey, Danielle Belgrave, Angela Fan, Ulrich Paquet, Jakub M. Tomczak, and Cheng Zhang, editors,Advances in Neural Information Processing Systems 38: 12 Annua...

    2024

  33. [33]

    Renrui Zhang, Dongzhi Jiang, Yichi Zhang, Haokun Lin, Ziyu Guo, Pengshuo Qiu, Aojun Zhou, Pan Lu, Kai-Wei Chang, Yu Qiao, Peng Gao, and Hongsheng Li. MATHVERSE: does your multi-modal LLM truly see the diagrams in visual math problems? In Ales Leonardis, Elisa Ricci, Stefan Roth, Olga Russakovsky, Torsten Sattler, and Gül Varol, editors,Computer Vision - E...

    2024

  34. [34]

    MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts

    Pan Lu, Hritik Bansal, Tony Xia, Jiacheng Liu, Chunyuan Li, Hannaneh Hajishirzi, Hao Cheng, Kai-Wei Chang, Michel Galley, and Jianfeng Gao. Mathvista: Evaluating math reasoning in visual contexts with gpt-4v, bard, and other large multimodal models.CoRR, abs/2310.02255, 2023

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  35. [35]

    Mmmu-pro: A more robust multi-discipline multimodal understanding benchmark

    Xiang Yue, Tianyu Zheng, Yuansheng Ni, Yubo Wang, Kai Zhang, Shengbang Tong, Yuxuan Sun, Botao Yu, Ge Zhang, Huan Sun, Yu Su, Wenhu Chen, and Graham Neubig. Mmmu-pro: A more robust multi-discipline multimodal understanding benchmark. In Wanxiang Che, Joyce Nabende, Ekaterina Shutova, and Mohammad Taher Pilehvar, editors,Proceedings of the 63rd Annual Meet...

    2025

  36. [36]

    Grok-1.5 vision preview

    xAI. Grok-1.5 vision preview. https://x.ai/news/grok-1.5v, 2024. Accessed: 2024-05- 20

    2024

  37. [37]

    Are we on the right way for evaluating large vision-language models? In Amir Globersons, Lester Mackey, Danielle Belgrave, Angela Fan, Ulrich Paquet, Jakub M

    Lin Chen, Jinsong Li, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Zehui Chen, Haodong Duan, Jiaqi Wang, Yu Qiao, Dahua Lin, and Feng Zhao. Are we on the right way for evaluating large vision-language models? In Amir Globersons, Lester Mackey, Danielle Belgrave, Angela Fan, Ulrich Paquet, Jakub M. Tomczak, and Cheng Zhang, editors,Advances in Neural Information P...

    2024

  38. [38]

    Simplevqa: Multimodal factuality evaluation for multimodal large language models.CoRR, abs/2502.13059, 2025

    Xianfu Cheng, Wei Zhang, Shiwei Zhang, Jian Yang, Xiangyuan Guan, Xianjie Wu, Xiang Li, Ge Zhang, Jiaheng Liu, Yuying Mai, Yutao Zeng, Zhoufutu Wen, Ke Jin, Baorui Wang, Weixiao Zhou, Yunhong Lu, Tongliang Li, Wenhao Huang, and Zhoujun Li. Simplevqa: Multimodal factuality evaluation for multimodal large language models.CoRR, abs/2502.13059, 2025

    work page arXiv 2025

  39. [39]

    Lucas Beyer, Andreas Steiner, André Susano Pinto, Alexander Kolesnikov, Xiao Wang, Daniel Salz, Maxim Neumann, Ibrahim Alabdulmohsin, Michael Tschannen, Emanuele Bugliarello, Thomas Unterthiner, Daniel Keysers, Skanda Koppula, Fangyu Liu, Adam Grycner, Alexey A. Gritsenko, Neil Houlsby, Manoj Kumar, Keran Rong, Julian Eisenschlos, Rishabh Kabra, Matthias ...

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  40. [40]

    Minesh Mathew, Viraj Bagal, Rubèn Tito, Dimosthenis Karatzas, Ernest Valveny, and C. V . Jawahar. Infographicvqa. InIEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022, Waikoloa, HI, USA, January 3-8, 2022, pages 2582–2591. IEEE, 2022

    2022

  41. [41]

    Manmatha, and C

    Minesh Mathew, Dimosthenis Karatzas, R. Manmatha, and C. V . Jawahar. Docvqa: A dataset for VQA on document images.CoRR, abs/2007.00398, 2020

    work page arXiv 2007

  42. [42]

    Joty, and Enamul Hoque

    Ahmed Masry, Do Xuan Long, Jia Qing Tan, Shafiq R. Joty, and Enamul Hoque. Chartqa: A benchmark for question answering about charts with visual and logical reasoning. In Smaranda Muresan, Preslav Nakov, and Aline Villavicencio, editors,Findings of the Association for Computational Linguistics: ACL 2022, Dublin, Ireland, May 22-27, 2022, Findings of ACL, p...

    2022

  43. [43]

    es se nt ia l

    Zirui Wang, Mengzhou Xia, Luxi He, Howard Chen, Yitao Liu, Richard Zhu, Kaiqu Liang, Xindi Wu, Haotian Liu, Sadhika Malladi, Alexis Chevalier, Sanjeev Arora, and Danqi Chen. 13 Charxiv: Charting gaps in realistic chart understanding in multimodal llms. In Amir Globersons, Lester Mackey, Danielle Belgrave, Angela Fan, Ulrich Paquet, Jakub M. Tomczak, and C...

    2024