arXiv AI
Adaptive Semantic Branching: Disagreement-Localized Cross-Trajectory Regeneration for Reasoning Candidate Expansion
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computation and Language
2d ago
Consolidating Reasoning with Test-Time Training
The paper introduces TTT‑Consolidate, a framework that leverages multi‑agent reasoning trajectories as adaptation data to improve subsequent reasoning. It consolidates these trajectories into parametric representations via test‑time training and then generates a new solution conditioned on the consolidated information, with meta‑learning used to optimize the consolidation process. Using the newly created OpenParallelThinking corpus of 12,810 math problems and trajectories from six LLMs, TTT‑Consolidate outperforms existing aggregation baselines by up to 20 percentage points on challenging math and STEM benchmarks, especially as model size grows.
arXiv AI
Aug 18
Divergent-Convergent Reasoning: Scaling Test-Time Compute through Structured Solution Synthesis
arXiv:2608. 15303v1 Announce Type: new Abstract: Test-time compute can substantially improve Large Language Model (LLM) reasoning performance, yet how and when additional compute helps remains poorly understood.
arXiv AI
Oct 9
Balancing Reference Guidance and Free Generation in Trajectory Rollouts for Reasoning RL
The paper introduces Adaptive Reference Guidance (ARG), a method that balances using a verified reference solution with free generation in trajectory rollouts for reasoning reinforcement learning. ARG selects a prefix of the reference to guide the model, continuing from it only if the resulting trajectory passes verification, otherwise falling back to the reference. Experiments on Qwen3-4B and Qwen3-8B across five mathematical reasoning benchmarks demonstrate that ARG achieves the highest aggregate pass@12 while maintaining competitive average sampled accuracy.
arXiv AI
Sep 1
To Retrieve or To Think? Cross-Boundary Context Evolution for Multi-hop Complex Reasoning
arXiv:2601.08747v3 Announce Type: replace-cross Abstract: Current context augmentation methods, such as retrieval-augmented generation, play a crucial role in bridging a model's internal knowledge bo...
arXiv Computation and Language
Sep 14
Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding
Chopthin-Consensus Power Sampling (CCPS) is a new inference-time decoding method for large language models that uses the Chopthin resampler to preserve diversity among particle trajectories. By enforcing an upper bound on weight ratios instead of equal-weight resampling, CCPS maintains a richer set of distinct reasoning paths and guarantees a lower bound on effective sample size. Coupled with a semantic-majority selection mechanism, CCPS achieves higher oracle coverage and matches or surpasses baseline accuracy on multiple reasoning benchmarks.
arXiv AI
Sep 10
Difficulty-Adaptive Tree-Structured Policy Optimization for Expanding Reasoning Coverage in RLVR
The paper introduces DATPO, a Difficulty‑Adaptive Sentence‑entropy‑guided Tree‑structured Policy Optimization method designed to improve reasoning coverage in Reinforcement Learning with Verifiable Rewards (RLVR). It builds on three design principles: adaptive difficulty rollouts, tree‑based rollouts, and sentence‑entropy‑guided forking to enhance semantic diversity. Experiments on mathematical reasoning benchmarks show that DATPO outperforms existing baselines, particularly in pass@k, leading to better test‑time scaling performance.