Mitigating Cognitive Inertia in Large Reasoning Models via Latent Spike Steering
While Large Reasoning Models (LRMs) have achieved remarkable performance by scaling test-time compute, they frequently suffer from Cognitive Inertia, a failure pattern manifesting as either overthinking (inertia of motion) or reasoning rigidity (inertia of direction). Existing detection methods, typically relying on superficial textual heuristics like self-correction tokens, often fail to capture the model's unvoiced internal conflicts. To address this, we propose STARS (Spike-Triggered Adaptive Reasoning Steering), a training-free framework designed to rectify cognitive inertia by monitoring latent dynamics. STARS identifies Cognitive Pivots-critical moments of reasoning transition-by detecting distinct L2 distance spikes in the hidden states. Upon detection, the framework employs geometric trajectory analysis to diagnose the structural nature of the transition and injects state-aware language cues to steer the model in real-time. Our experiments across diverse benchmarks confirm that STARS efficiently curtails redundant loops while improving accuracy through the adaptive correction of erroneous trajectories. STARS offers a robust, unsupervised mechanism to optimize the reasoning process of LRMs without requiring additional fine-tuning.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
MIND: From Passive Mimicry to Active Reasoning through Capability-Aware Multi-Perspective CoT Distillation
While Large Language Models (LLMs) have emerged with remarkable capabilities in complex tasks through Chain-of-Thought reasoning, practical resource constraints have sparked interest in transferring these abilities to sm…
Domain GeneralizationSelf-Awareness before Action: Mitigating Logical Inertia via Proactive Cognitive Awareness
Large language models perform well on many reasoning tasks, yet they often lack awareness of whether their current knowledge or reasoning state is complete. In non-interactive puzzle settings, the narrative is fixed and …
Bi-directional CognitiveThinking Network for Machine Reading Comprehension
We propose a novel Bi-directional Cognitive Knowledge Framework (BCKF) for reading comprehension from the perspective of complementary learning systems theory. It aims to simulate two ways of thinking in the brain to ans…
Machine Reading ComprehensionReading ComprehensionBi-directional Cognitive Thinking Network for Machine Reading Comprehension
We propose a novel Bi-directional Cognitive Knowledge Framework (BCKF) for reading comprehension from the perspective of complementary learning systems theory. It aims to simulate two ways of thinking in the brain to ans…
Machine Reading ComprehensionReading ComprehensionAttention at Rest Stays at Rest: Breaking Visual Inertia for Cognitive Hallucination Mitigation
Like a body at rest that stays at rest, we find that visual attention in multimodal large language models (MLLMs) exhibits pronounced inertia, remaining largely static once settled during early decoding steps and failing…