Pith. sign in

REVIEW 19 cited by

Boosting the Generalization and Reasoning of Vision Language Models with Curriculum Reinforcement Learning

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.07065 v1 pith:Z7NUCBYK submitted 2025-03-10 cs.CV

Huilin Deng , Ding Zou , Rui Ma , Hongchen Luo , Yang Cao , Yu Kang This is my paper
classification cs.CV
keywords modelscurr-reftlearningreinforcementvlmscapabilitiescurriculumlanguage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

While state-of-the-art vision-language models (VLMs) have demonstrated remarkable capabilities in complex visual-text tasks, their success heavily relies on massive model scaling, limiting their practical deployment. Small-scale VLMs offer a more practical alternative but face significant challenges when trained with traditional supervised fine-tuning (SFT), particularly in two aspects: out-of-domain (OOD) generalization and reasoning abilities, which significantly lags behind the contemporary Large language models (LLMs). To address these challenges, we propose Curriculum Reinforcement Finetuning (Curr-ReFT), a novel post-training paradigm specifically designed for small-scale VLMs. Inspired by the success of reinforcement learning in LLMs, Curr-ReFT comprises two sequential stages: (1) Curriculum Reinforcement Learning, which ensures steady progression of model capabilities through difficulty-aware reward design, transitioning from basic visual perception to complex reasoning tasks; and (2) Rejected Sampling-based Self-improvement, which maintains the fundamental capabilities of VLMs through selective learning from high-quality multimodal and language examples. Extensive experiments demonstrate that models trained with Curr-ReFT paradigm achieve state-of-the-art performance across various visual tasks in both in-domain and out-of-domain settings. Moreover, our Curr-ReFT enhanced 3B model matches the performance of 32B-parameter models, demonstrating that efficient training paradigms can effectively bridge the gap between small and large models.

Discussion (0). Sign in to comment.

Forward citations

Cited by 19 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Towards One-to-Many Temporal Grounding

    cs.CV 2026-06 unverdicted novelty 7.0 of 10

    Introduces OMTG benchmark with C-Acc and EtF1 metrics, a 56k dataset, and caption/temporal rewards, reaching 43.65% EtF1 SOTA on the new bench.

  2. S-GRPO: Unified Post-Training for Large Vision-Language Models

    cs.LG 2026-04 unverdicted novelty 7.0 of 10

    S-GRPO unifies SFT and RL for LVLMs via conditional ground-truth injection that supplies a maximal-reward anchor when group exploration fails completely.

  3. Dense Reward for Multi-View 3D Reasoning with Global Maps and Local Views

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    DR-MV3D decomposes MV3D-VQA into global map construction, question-conditioned view planning, and egocentric grounding, supervised by global consistency and local trajectory rewards optimized via GRPO.

  4. RadThinking: A Dataset for Longitudinal Clinical Reasoning in Radiology

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    RadThinking releases a large longitudinal CT VQA dataset stratified into foundation perception questions, single-rule reasoning questions, and compositional multi-step chains grounded in clinical reporting standards f...

  5. DARE: Difficulty-Adaptive Reinforcement Learning with Co-Evolved Difficulty Estimation

    cs.LG 2026-05 unverdicted novelty 6.0 of 10

    DARE co-evolves difficulty estimation and policy in RL for LLMs to improve training efficiency, final performance, and inference speed by using tailored strategies for different difficulty levels.

  6. RL-RIG: A Generative Spatial Reasoner via Intrinsic Reflection

    cs.CV 2026-02 unverdicted novelty 6.0 of 10

    RL-RIG uses a generate-reflect-edit loop with reinforcement learning to improve spatial accuracy in image generation, reporting up to 11% gains over prior open-source models on scene-graph metrics.

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

    cs.CV 2025-04 unverdicted novelty 6.0 of 10

    VLM-R1 applies R1-style RL using rule-based rewards on visual tasks with clear ground truth to achieve competitive performance and superior generalization over SFT in vision-language models.

  8. From Structure to Synergy: A Survey of Vision-Language Perception Paradigm Evolution in Multimodal Large Language Models

    cs.CL 2026-06 unverdicted novelty 5.0 of 10

    The survey formalizes MLLM perception as a unified vision-language capability and traces its evolution via a new five-stage taxonomy while outlining future challenges.

  9. D$^2$Evo: Dual Difficulty-Aware Self-Evolution for Data-Efficient Reinforcement Learning

    cs.LG 2026-05 unverdicted novelty 5.0 of 10

    D²Evo mines medium-difficulty anchors from the current model, trains a Questioner to generate matching questions, and jointly optimizes Solver and Questioner for progressive gains, outperforming baselines on math reas...

  10. Rethinking Reward Signals in Video GRPO: When Scores Become Targets

    cs.CV 2025-11 reject novelty 5.0 of 10

    A self-paced GRPO variant that adaptively reweights visual, temporal, and text-alignment reward components as the generator improves reports small VBench gains over static-reward baselines.

  11. VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning

    cs.CV 2025-04 unverdicted novelty 5.0 of 10

    Reinforcement fine-tuning with temporal rewards produces VideoChat-R1, a video MLLM showing large gains on spatio-temporal perception benchmarks such as +31.8 temporal grounding and +31.2 object tracking.

  12. Stage-1 Controls the Entropy Regime, Not the Outcome

    cs.LG 2026-06 unverdicted novelty 4.0 of 10

    Stage-1 warm-starts control the entropy regime entering RL but yield small and localized effects on final in-domain and out-of-domain performance.

  13. S-GRPO: Unified Post-Training for Large Vision-Language Models

    cs.LG 2026-04 conditional novelty 4.0 of 10

    Injecting a max-reward ground-truth answer into a GRPO group whenever all sampled responses fail prevents optimization collapse and matches full fine-tuning's domain scores on three benchmark tasks.

  14. Curr-RLCER:Curriculum Reinforcement Learning For Coherence Explainable Recommendation

    cs.IR 2026-04 unverdicted novelty 4.0 of 10

    Curr-RLCER applies curriculum reinforcement learning with coherence-driven rewards to align generated explanations with predicted ratings in explainable recommendation systems.

  15. Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning

    cs.CV 2025-07 unverdicted novelty 4.0 of 10

    A pipeline of chain-of-thought data synthesis, LoRA-based supervised fine-tuning, rejection sampling, and rule-based reinforcement learning raises multi-image grounding accuracy by 9.04% on MIG-Bench and 4.41% on aver...

  16. Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle

    cs.CL 2025-09 conditional novelty 3.0 of 10

    A survey that maps reinforcement learning methods, datasets, benchmarks, and open-source tools across the full training lifecycle of large language models, focusing on verifiable-reward reasoning.

  17. Empowering Multimodal LLMs with External Tools: A Comprehensive Survey

    cs.CV 2025-08 unverdicted novelty 2.0 of 10

    A survey paper maps how external tools are used to augment multimodal large language models across data, tasks, evaluation, and future directions.

  18. Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey

    cs.CV 2025-03 unverdicted novelty 2.0 of 10

    The paper provides the first comprehensive survey of multimodal chain-of-thought reasoning, including foundational concepts, a taxonomy of methodologies, application analyses, challenges, and future directions.

  19. Aligning Large Vision-Language Models by Deep Reinforcement Learning and Direct Preference Optimization

    cs.LG 2025-09 unverdicted

    A survey categorizing deep reinforcement learning and direct preference optimization methods for aligning large vision-language models, with tables of studies and datasets and no new experimental result.

Pith tools