Chain-of-thought (CoT) often improves multi-step reasoning, but it remains unclear what kind of additional sequential computation longer traces actually enable. We connect CoT to Bennett’s logical depth, separating an answer’s description length from the sequential effort required to derive it, and view a CoT budget of T steps as a qualitative cap on realizable sequential computation. To operationalize realized depth beyond raw length, we introduce Effective Logical Depth (ELD), a deletion-based measure of step necessity under a specified inference interface. Across depth-controlled prefix-sum tasks and GSM8K rationale perturbations, we observe two consistent signatures of a Time-for-Accuracy tradeoff: (i) plateau-to-transition accuracy curves as the budget increases from being below to matching the task’s required depth, and (ii) sparse, position-dependent deletion sensitivity concentrated in early steps for deeper instances. On GSM8K, an Extract interface, where the model reads off the answer from the remaining rationale, remains near-perfect even after prefix deletions, whereas a Repair interface, where the model must re-solve from truncated rationale context, degrades markedly. Moreover, Socratic human rationales are consistently more robust than Main rationales under Repair. These results suggest that longer CoT helps primarily when it enables additional effective sequential computation, and that deletion-based diagnostics can distinguish computational steps from redundant ones.

Despite extensive safety alignment, Large Language Models (LLMs) often fail against jailbreak attacks. While machine unlearning has emerged as a promising defense by erasing specific harmful parameters, current methods remain vulnerable to diverse jailbreaks. We first conduct an empirical study and discover that this failure mechanism is caused by jailbreaks primarily activating non-erased parameters in the intermediate layers. Further, by probing the underlying mechanism through which these circumvented parameters reassemble into the prohibited output, we verify the persistent existence of dynamic jailbreak paths and show that the inability to rectify them constitutes the fundamental gap in existing unlearning defenses. To bridge this gap, we propose Jailbreak Path Unlearning (JPU), which is the first to rectify dynamic jailbreak paths towards safety anchors by dynamically mining on-policy adversarial samples to expose vulnerabilities and identify jailbreak paths. Extensive experiments demonstrate that JPU significantly enhances jailbreak resistance against dynamic attacks while preserving the model’s utility.

With the surge of online misinformation, Large Language Models (LLMs) and Reasoning Large Language Models (RLMs) serving as Automatic Fact-Checking (AFC) systems have emerged as a prominent paradigm for reliable, explainable verification. However, our empirical study reveals that this paradigm faces a critical risk asymmetry challenge when deployed in real-world under resource-constrained environments. While Hotspot Perception Ability (HPA), the capacity to dynamically allocate reasoning resources based on social impact, is essential to mitigate this risk, existing benchmarks lack the social metadata and evaluation framework to meet this urgent evaluation needs, thereby hindering the advancement of these AFC systems. To bridge this gap, we introduce TrendFact, the first benchmark capable of evaluating HPA and three fact-checking tasks. It consists of 7,643 curated samples sourced from trending platforms and professional datasets, with an evidence library containing 366,634 entries. To enable HPA assessment, we propose two novel metrics: the Explanation Consistency Score (ECS) to evaluate the reliability of verification reasoning, and the Hotspot Claim Perception Index (HCPI) to quantify the overall HPA of AFC systems. Extensive experiments demonstrate that existing AFC systems exhibit limited performance on TrendFact. Furthermore, our proposed FactISR framework effectively enhances HPA and computational efficiency for RLM-driven systems.

Large language models (LLMs) generate human-aligned content under certain safety constraints. However, the current known technique “jailbreak prompt” can circumvent safety-aligned measures and induce LLMs to output malicious content. Research on Jailbreaking can help identify vulnerabilities in LLMs and guide the development of robust security frameworks. To circumvent the issue of attack templates becoming obsolete as models evolve, existing methods adopt iterative mutation and dynamic optimization to facilitate more automated jailbreak attacks. However, these methods face two challenges: inefficiency and repetitive optimization, as they overlook the value of past attack experiences. To better integrate past attack experiences to assist current jailbreak attempts, we propose the JailExpert, an automated jailbreak framework, which is the first to achieve a formal representation of experience structure, group experiences based on semantic drift, and support the dynamic updating of the experience pool. Extensive experiments demonstrate that JailExpert significantly improves both attack effectiveness and efficiency. Compared to the current state-of-the-art black-box jailbreak method, JailExpert achieves an average increase of 24% in attack success rate and 2.7 times improvement in attack efficiency.

Retrieval-Augmented Large Language Models(RALMs) have made significant strides in enhancing the accuracy of generated responses. However, existing research often overlooks the data quality issues within retrieval results, often caused by inaccurate existing vector-distance-based retrieval methods. We propose to boost the precision of RALMs’ answers from a data quality perspective through the Context-Driven Index Trimming (CDIT) framework, where Context Matching Dependencies (CMDs) are employed as logical data quality rules to capture and regulate the consistency between retrieved contexts. Based on the semantic comprehension capabilities of Large Language Models (LLMs), CDIT can effectively identify and discard retrieval results that are inconsistent with the query context and further modify indexes in the database, thereby improving answer quality. Experiments demonstrate average improvement of 3.75% in accuracy on challenging open-domain question-answering tasks. Also, the flexibility of CDIT is verified through its compatibility with various language models and indexing methods, which offers a promising approach to bolster RALMs’ data quality and retrieval precision jointly.