ZPPO improves distillation to small vision-language models by using binary and negative candidate prompts plus a replay buffer for hard questions, outperforming standard distillation and GRPO on a 31-benchmark suite with largest gains at the 0.8B scale.
Mmr1: Enhancing multimodal reasoning with variance-aware sampling and open resources
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
roles
baseline 1polarities
baseline 1representative citing papers
Reasoning VLMs show lower robustness to semantic visual distractions than to perceptual corruptions, with distractions entering their reasoning chains and causing errors.
This survey introduces the Generate-Filter-Control-Replay (GFCR) taxonomy to structure rollout pipelines for RL-based post-training of reasoning LLMs.
TRON supplies 520 rule-verifiable online visual reasoning environments across five ability buckets that generate unlimited training instances for RL post-training, yielding consistent gains on ten external multimodal benchmarks for three vision-language models.
AnE combines Truth Anchor Expansion and Scaffold-Stripping to deliver 10.3% gains on eight multimodal reasoning benchmarks for MLLMs.
Staged post-training that first solidifies visual perception before visual and textual reasoning improves VLM accuracy and shortens reasoning traces on visual math and perception benchmarks.
PRISM adds a black-box on-policy distillation stage with an MoE discriminator between SFT and RLVR for multimodal models, yielding +4.4 and +6.0 average accuracy gains on 4B and 8B Qwen3-VL models over the standard baseline.
CHAI framework pairs AI pre-captions with expert human critiques to produce precise video descriptions, enabling open models to outperform closed ones like Gemini-3.1-Pro and improve fine-grained control in video generation models.
LongVT adds native video-cropping tool calling to LMMs for interleaved multimodal chain-of-tool-thought reasoning on long videos and releases VideoSIAH data for training and evaluation.
Activation Replay boosts multimodal reasoning in post-trained LMMs by replaying low-entropy activations from base models to RLVR counterparts at test time via visual token manipulation.
LASER uses Visual Grounding Reward and Sink Suppression Reward to preserve visual attention trajectories and suppress sink tokens, reducing visual forgetting in LVLMs.
EVA generates adaptive Latent_slot tokens as internal visual thoughts, trained end-to-end with text tokens via D-GSPO on the EVA-230K dataset, claiming performance gains and better inference efficiency.
VeriEvol decouples prompt difficulty evolution from answer reliability verification to scale verified data for visual math reasoning, lifting benchmark accuracy from 35.42 to 54.73 and adding +3.88 in GRPO RL.
citing papers explorer
-
Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients
ZPPO improves distillation to small vision-language models by using binary and negative candidate prompts plus a replay buffer for hard questions, outperforming standard distillation and GRPO on a 31-benchmark suite with largest gains at the 0.8B scale.
-
Are Reasoning Vision-Language Models Robust to Semantic Visual Distractions?
Reasoning VLMs show lower robustness to semantic visual distractions than to perceptual corruptions, with distractions entering their reasoning chains and causing errors.
-
Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning
This survey introduces the Generate-Filter-Control-Replay (GFCR) taxonomy to structure rollout pipelines for RL-based post-training of reasoning LLMs.
-
TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL
TRON supplies 520 rule-verifiable online visual reasoning environments across five ability buckets that generate unlimited training instances for RL post-training, yielding consistent gains on ten external multimodal benchmarks for three vision-language models.
-
AnE: Pushing the Reasoning Frontier of Multimodal LLMs via Anchor Evolution
AnE combines Truth Anchor Expansion and Scaffold-Stripping to deliver 10.3% gains on eight multimodal reasoning benchmarks for MLLMs.
-
From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models
Staged post-training that first solidifies visual perception before visual and textual reasoning improves VLM accuracy and shortens reasoning traces on visual math and perception benchmarks.
-
Beyond SFT-to-RL: Pre-alignment via Black-Box On-Policy Distillation for Multimodal RL
PRISM adds a black-box on-policy distillation stage with an MoE discriminator between SFT and RLVR for multimodal models, yielding +4.4 and +6.0 average accuracy gains on 4B and 8B Qwen3-VL models over the standard baseline.
-
Building a Precise Video Language with Human-AI Oversight
CHAI framework pairs AI pre-captions with expert human critiques to produce precise video descriptions, enabling open models to outperform closed ones like Gemini-3.1-Pro and improve fine-grained control in video generation models.
-
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling
LongVT adds native video-cropping tool calling to LMMs for interleaved multimodal chain-of-tool-thought reasoning on long videos and releases VideoSIAH data for training and evaluation.
-
Boosting Reasoning in Large Multimodal Models via Activation Replay
Activation Replay boosts multimodal reasoning in post-trained LMMs by replaying low-entropy activations from base models to RLVR counterparts at test time via visual token manipulation.
-
LASER: A Corrective Lens for LVLMs via Visual Attention Preservation and Sink Suppression
LASER uses Visual Grounding Reward and Sink Suppression Reward to preserve visual attention trajectories and suppress sink tokens, reducing visual forgetting in LVLMs.
-
Latent Visual States for Efficient Multimodal Reasoning
EVA generates adaptive Latent_slot tokens as internal visual thoughts, trained end-to-end with text tokens via D-GSPO on the EVA-230K dataset, claiming performance gains and better inference efficiency.
-
VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct
VeriEvol decouples prompt difficulty evolution from answer reliability verification to scale verified data for visual math reasoning, lifting benchmark accuracy from 35.42 to 54.73 and adding +3.88 in GRPO RL.