Behavior Forecasters trained on LRM trajectories outperform larger models in predicting repeatability and input sensitivity at low cost.
arXiv preprint arXiv:2405.15092 , year=
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
ReasoningFlow represents LLM reasoning traces as DAGs, finding structural similarity across models and that most erroneous steps are unused in final answers.
DOLORES, an agent using a formal language for meta-reasoning to construct adaptive scaffolds on the fly, outperforms prior scaffolding methods by 24.8% on average across four hard benchmarks and multiple model sizes.
CIE-Scorer detects unfaithful CoT by tracing compact sentence-level circuits, building internal-external reasoning graphs, and scoring their discrepancy with Fused Gromov-Wasserstein distance, reporting SOTA results on FaithCoT-Bench with reduced circuit cost.
Introduces EPC-AW to mitigate epistemic miscalibration in LLM multi-agent planning via consistency-based selection and refinement, reporting 9.75% average success improvement.
Chain-of-thought steps in LLMs vary in causal influence; many are decorative, TTS identifies them, and a latent steering direction can switch whether a model 'thinks' through a step.
SCPRM adds prefix conditioning and schema distance to process reward models so that Monte Carlo Tree Search can explore knowledge-graph reasoning paths with both cumulative and future guidance, yielding a 1.18% average Hits@k gain on medical, legal, and CWQ tasks.
citing papers explorer
-
Forecasting Future Behavior as a Learning Task
Behavior Forecasters trained on LRM trajectories outperform larger models in predicting repeatability and input sensitivity at low cost.
-
ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces
ReasoningFlow represents LLM reasoning traces as DAGs, finding structural similarity across models and that most erroneous steps are unused in final answers.
-
Deep Reasoning in General Purpose Agents via Structured Meta-Cognition
DOLORES, an agent using a formal language for meta-reasoning to construct adaptive scaffolds on the fly, outperforms prior scaffolding methods by 24.8% on average across four hard benchmarks and multiple model sizes.
-
Detecting Unfaithful Chain-of-Thought via Circuit-Guided Internal-External Discrepancy
CIE-Scorer detects unfaithful CoT by tracing compact sentence-level circuits, building internal-external reasoning graphs, and scoring their discrepancy with Fused Gromov-Wasserstein distance, reporting SOTA results on FaithCoT-Bench with reduced circuit cost.
-
When Planning Fails Despite Correct Execution: On Epistemic Calibration for LLM-Based Multi-Agent Systems
Introduces EPC-AW to mitigate epistemic miscalibration in LLM multi-agent planning via consistency-based selection and refinement, reporting 9.75% average success improvement.
-
Can Aha Moments Be Fake? Towards Quantifying Decorative and True Thinking in Chain-of-Thought
Chain-of-thought steps in LLMs vary in causal influence; many are decorative, TTS identifies them, and a latent steering direction can switch whether a model 'thinks' through a step.
-
SCPRM: A Schema-aware Cumulative Process Reward Model for Knowledge Graph Question Answering
SCPRM adds prefix conditioning and schema distance to process reward models so that Monte Carlo Tree Search can explore knowledge-graph reasoning paths with both cumulative and future guidance, yielding a 1.18% average Hits@k gain on medical, legal, and CWQ tasks.