DiG-Plan decouples combinatorial tool-set exploration via diffusion from structural refinement via autoregressive decoding, raising Pass@10 coverage from 0.32 to 0.94 in controlled tests and improving benchmark performance by 10%.
arXiv preprint arXiv:2502.05078 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
SCA applies the Information Bottleneck principle via NIBS and GIBS methods to identify erroneous steps in black-box LLM reasoning and boosts self-correction success by up to 13.5%.
citing papers explorer
-
DiG-Plan: Mitigating Early Commitment for Tool-Graph Planning via Diffusion Guidance
DiG-Plan decouples combinatorial tool-set exploration via diffusion from structural refinement via autoregressive decoding, raising Pass@10 coverage from 0.32 to 0.94 in controlled tests and improving benchmark performance by 10%.
-
Diagnosing Multi-step Reasoning Failures in Black-box LLMs via Stepwise Confidence Attribution
SCA applies the Information Bottleneck principle via NIBS and GIBS methods to identify erroneous steps in black-box LLM reasoning and boosts self-correction success by up to 13.5%.