pith. sign in

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

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CV 1

years

2026 1

verdicts

UNVERDICTED 1

representative citing papers

Reinforcement Learning with Robust Rubric Rewards

cs.CV · 2026-05-28 · unverdicted · novelty 5.0

RLR³ extends RLVR to criterion-level rubric verification via dual execution paths, minimal exposure masking, hierarchical aggregation, and saturation mitigation, delivering 4.7-point gains over base on 15 benchmarks with Qwen3-VL-30B-A3B.

citing papers explorer

Showing 1 of 1 citing paper.

  • Reinforcement Learning with Robust Rubric Rewards cs.CV · 2026-05-28 · unverdicted · none · ref 29

    RLR³ extends RLVR to criterion-level rubric verification via dual execution paths, minimal exposure masking, hierarchical aggregation, and saturation mitigation, delivering 4.7-point gains over base on 15 benchmarks with Qwen3-VL-30B-A3B.