Maven is an RL method using answer-conditioned evidence-state values to assign rewards to add, link, and drop actions on evidence memory, outperforming outcome-only baselines on LongBench v2, LongReason, and RULER.
Qwenlong-l1: Towards long-context large reasoning models with reinforcement learning
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 1
dataset 1
citation-polarity summary
representative citing papers
StoryReward, trained on a new 100k story preference dataset, sets state-of-the-art performance on the introduced StoryRMB benchmark for aligning LLM stories with human preferences.
Synthetic page-ranked reasoning traces plus low-strength model merging give a 32B VLM 58.3 on MMLongBenchDoc, beating a 235B teacher while cutting output tokens ~12× versus explicit reasoning.
OPSDL improves long-context LLM performance by having the model self-distill from its short-context capability using point-wise reverse KL divergence on generated tokens, outperforming SFT and DPO on benchmarks without harming short-context abilities.
Combines GRPO with teacher-guided on-policy distillation and introduces LongBlocks dataset to yield more stable long-context reasoning than either method alone.
LongAct uses saliency from high-magnitude activations to guide sparse weight updates in long-context RL, yielding about 8% gains on LongBench v2 across multiple algorithms.
Decomposing long-context reasoning into atomic skills, synthesizing targeted pseudo-datasets, and applying RL improves LLM performance on long-context benchmarks by an average of 7.7%.
GLM-4.5, a 355B-parameter MoE model with hybrid reasoning, scores 70.1% on TAU-Bench, 91.0% on AIME 24, and 64.2% on SWE-bench Verified while ranking 3rd overall and 2nd on agentic benchmarks.
citing papers explorer
-
Evidence-State Rewards for Long-Context Reasoning
Maven is an RL method using answer-conditioned evidence-state values to assign rewards to add, link, and drop actions on evidence memory, outperforming outcome-only baselines on LongBench v2, LongReason, and RULER.
-
StoryAlign: Evaluating and Training Reward Models for Story Generation
StoryReward, trained on a new 100k story preference dataset, sets state-of-the-art performance on the introduced StoryRMB benchmark for aligning LLM stories with human preferences.
-
Internalized Reasoning for Long-Context Visual Document Understanding
Synthetic page-ranked reasoning traces plus low-strength model merging give a 32B VLM 58.3 on MMLongBenchDoc, beating a 235B teacher while cutting output tokens ~12× versus explicit reasoning.
-
OPSDL: On-Policy Self-Distillation for Long-Context Language Models
OPSDL improves long-context LLM performance by having the model self-distill from its short-context capability using point-wise reverse KL divergence on generated tokens, outperforming SFT and DPO on benchmarks without harming short-context abilities.
-
A Recipe for Long-Context Reasoning in Large Language Models via On-Policy Optimization and Distillation
Combines GRPO with teacher-guided on-policy distillation and introduces LongBlocks dataset to yield more stable long-context reasoning than either method alone.
-
LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning
LongAct uses saliency from high-magnitude activations to guide sparse weight updates in long-context RL, yielding about 8% gains on LongBench v2 across multiple algorithms.
-
A Decomposition Perspective to Long-context Reasoning for LLMs
Decomposing long-context reasoning into atomic skills, synthesizing targeted pseudo-datasets, and applying RL improves LLM performance on long-context benchmarks by an average of 7.7%.
-
GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models
GLM-4.5, a 355B-parameter MoE model with hybrid reasoning, scores 70.1% on TAU-Bench, 91.0% on AIME 24, and 64.2% on SWE-bench Verified while ranking 3rd overall and 2nd on agentic benchmarks.
- MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent