REVIEW 10 cited by
Not All Errors Are Created Equal: ASCoT Addresses Late-Stage Fragility in Efficient LLM Reasoning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Not All Errors Are Created Equal: ASCoT Addresses Late-Stage Fragility in Efficient LLM Reasoning
read the original abstract
While Chain-of-Thought (CoT) prompting empowers Large Language Models (LLMs), ensuring reasoning reliability remains an open challenge. Contrary to the prevailing cascading failure hypothesis which posits that early errors are most detrimental, we identify a counter-intuitive phenomenon termed \textbf{Late-Stage Fragility}: errors introduced in later reasoning stages are significantly more prone to corrupting final answers. To address this, we introduce ASCoT (Adaptive Self-Correction Chain-of-Thought), a method harmonizing efficiency with robust verification. ASCoT first employs semantic pruning to compress redundant steps, then utilizes an Adaptive Verification Manager (AVM) to prioritize high risk, late-stage steps via a positional impact score, triggering a Multi-Perspective Self-Correction Engine (MSCE) only when necessary. Experiments on GSM8K and MATH-500 demonstrate that ASCoT effectively reallocates computational resources: it reduces token usage by 21\%--30\% for LLaMA-3.1-8B with negligible accuracy drops ($<1.8\%$), achieving a superior trade-off between inference efficiency and reasoning fidelity.
Forward citations
Cited by 10 Pith papers
-
Fragile Thoughts: How Large Language Models Handle Chain-of-Thought Perturbations
LLMs show heterogeneous robustness to five types of chain-of-thought perturbations, with MathError causing 50-60% accuracy loss in small models but scaling benefits, UnitConversion remaining hard across sizes, and Ext...
-
Parameter-Efficient Fine-Tuning for Spiking Point Cloud Models
SpikePEFT matches or beats full fine-tuning on spiking point-cloud benchmarks while updating only ~5% of parameters via neuron-dynamics and silent-state adapters.
-
FADE: Mitigating Hallucinations by Reducing Language-Prior Dominance in Large Vision-Language Models
FADE attenuates FFN outputs in LVLMs based on layer-wise information flow analysis to mitigate hallucinations, shown effective on POPE, CHAIR, and MME benchmarks.
-
FADE: Mitigating Hallucinations by Reducing Language-Prior Dominance in Large Vision-Language Models
Attenuating FFN outputs at mid-to-late transformer layers reduces language-prior dominance and thereby mitigates object hallucinations in LVLMs while preserving efficiency.
-
CFMS: A Coarse-to-Fine Multimodal Synthesis Framework for Enhanced Tabular Reasoning
CFMS is a coarse-to-fine framework that uses MLLMs to create a multi-perspective knowledge tuple as a reasoning map for symbolic table operations, yielding competitive accuracy on WikiTQ and TabFact.
-
Better Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning
Warm-starting on-policy self-distillation with a correctness-filtered, prompt-switched SFT bootstrap improves compression and accuracy of reasoning traces across math benchmarks.
-
FADE: Mitigating Hallucinations by Reducing Language-Prior Dominance in Large Vision-Language Models
FADE attenuates FFN outputs at critical layers in LVLMs to curb language-prior dominance and cut hallucinations, shown effective on POPE, CHAIR, and MME across three models.
-
Sparsity-Aware Voxel Attention and Foreground Modulation for 3D Semantic Scene Completion
VoxSAMNet introduces sparsity-aware deformable attention via a dummy node and foreground modulation with dropout plus text-guided filtering to reach new state-of-the-art mIoU of 18.2% on SemanticKITTI and 20.2% on SSC...
-
Controllable Singing Style Conversion with Boundary-Aware Information Bottleneck
A singing voice conversion system with boundary-aware information bottleneck and high-frequency augmentation achieves the best naturalness in SVCC2025 subjective tests while using less extra data than competitors.
-
Agentic Reasoning for Large Language Models
The survey structures agentic reasoning for LLMs into foundational, self-evolving, and collective multi-agent layers while distinguishing in-context orchestration from post-training optimization and reviewing applicat...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.