Static analysis is a powerful technique for bug detection in critical systems like operating system kernels. However, designing and implementing static analyzers is challenging, time-consuming, and typically limited to predefined bug patterns. While large language models (LLMs) have shown promise for static analysis, directly applying them to scan large codebases remains impractical due to computational constraints and contextual limitations. We present KNighter, the first approach that unlocks practical LLM-based static analysis by automatically synthesizing static analyzers from historical bug patterns. Rather than using LLMs to directly analyze massive codebases, our key insight is leveraging LLMs to generate specialized static analyzers guided by historical patch knowledge. KNighter implements this vision through a multi-stage synthesis pipeline that validates checker correctness against original patches and employs an automated refinement process to iteratively reduce false positives. Our evaluation on the Linux kernel demonstrates that KNighter generates high-precision checkers capable of detecting diverse bug patterns overlooked by existing human-written analyzers. To date, KNighter-synthesized checkers have discovered 70 new bugs/vulnerabilities in the Linux kernel, with 56 confirmed and 41 already fixed. 11 of these findings have been assigned CVE numbers. This work establishes an entirely new paradigm for scalable, reliable, and traceable LLM-based static analysis for real-world systems via checker synthesis.

Quadrupeds have gained rapid advancement in their capability of traversing across complex terrains. The adoption of deep Reinforcement Learning (RL), transformers and various knowledge transfer techniques can greatly reduce the sim-to-real gap. However, the classical teacher-student framework commonly used in existing locomotion policies requires a pre-trained teacher and leverages the privilege information to guide the student policy. With the implementation of large-scale models in robotics controllers, especially transformers-based ones, this knowledge distillation technique starts to show its weakness in efficiency, due to the requirement of multiple supervised stages. In this paper, we propose Unified Locomotion Transformer (ULT), a new transformer-based framework to unify the processes of knowledge transfer and policy optimization in a single network while still taking advantage of privilege information. The policies are optimized with reinforcement learning, next state-action prediction, and action imitation, all in just one training stage, to achieve zero-shot deployment. Evaluation results demonstrate that with ULT, optimal teacher and student policies can be obtained at the same time, greatly easing the difficulty in knowledge transfer, even with complex transformer-based models.

Large language models (LLMs) have shown great promise as language understanding and decision making tools, and they have permeated various aspects of our everyday life. However, their widespread availability also comes with novel risks, such as generating harmful, unethical, or offensive content, via an attack called jailbreaking. Despite extensive efforts from LLM developers to align LLMs using human feedback, they are still susceptible to jailbreak attacks. To tackle this issue, researchers often employ red-teaming to understand and investigate jailbreak prompts. However, existing red-teaming approaches lack effectiveness, scalability, or both. To address these issues, we propose JBFuzz, a novel effective, automated, and scalable red-teaming technique for jailbreaking LLMs. JBFuzz is inspired by the success of fuzzing for detecting bugs/vulnerabilities in software. We overcome three challenges related to effectiveness and scalability by devising novel seed prompts, a lightweight mutation engine, and a lightweight and accurate evaluator for guiding the fuzzer. Assimilating all three solutions results in a potent fuzzer that only requires black-box access to the target LLM. We perform extensive experimental evaluation of JBFuzz using nine popular and widely-used LLMs. We find that JBFuzz successfully jailbreaks all LLMs for various harmful/unethical questions, with an average attack success rate of 99%. We also find that JBFuzz is extremely efficient as it jailbreaks a given LLM for a given question in 60 seconds on average. Our work highlights the susceptibility of the state-of-the-art LLMs to jailbreak attacks even after safety alignment, and serves as a valuable red-teaming tool for LLM developers.

The remarkable achievements of large language models (LLMs) have led many to conclude that they exhibit a form of intelligence. This is as opposed to explanations of their capabilities based on their ability to perform relatively simple manipulations of vast volumes of data. To illuminate the distinction between these explanations, we introduce a novel generative model that generates tokens on the basis of human interpretable concepts represented as latent discrete variables. Under mild conditions, even when the mapping from the latent space to the observed space is non-invertible, we establish an identifiability result: the representations learned by LLMs through next-token prediction can be approximately modeled as the logarithm of the posterior probabilities of these latent discrete concepts, up to an invertible linear transformation. This theoretical finding not only provides evidence that LLMs capture underlying generative factors, but also strongly reinforces the linear representation hypothesis, which posits that LLMs learn linear representations of human-interpretable concepts. Empirically, we validate our theoretical results through evaluations on both simulation data and the Pythia, Llama, and DeepSeek model families.

The integration of Agentic AI into scientific discovery marks a new frontier in research automation. These AI systems, capable of reasoning, planning, and autonomous decision-making, are transforming how scientists perform literature review, generate hypotheses, conduct experiments, and analyze results. This survey provides a comprehensive overview of Agentic AI for scientific discovery, categorizing existing systems and tools, and highlighting recent progress across fields such as chemistry, biology, and materials science. We discuss key evaluation metrics, implementation frameworks, and commonly used datasets to offer a detailed understanding of the current state of the field. Finally, we address critical challenges, such as literature review automation, system reliability, and ethical concerns, while outlining future research directions that emphasize human-AI collaboration and enhanced system calibration. The rapid advancements of Large Language Models (LLMs) (Touvron et al., 2023; Anil et al., 2023; Achiam et al., 2023) have opened a new era in scientific discovery, with Agentic AI systems (Kim et al., 2024; Guo et al., 2023; Wang et al., 2024; Abramovich et al., 2024) emerging as powerful tools for automating complex research workflows. Unlike traditional AI, Agentic AI systems are designed to operate with a high degree of autonomy, allowing them to independently perform tasks such as hypothesis generation, literature review, experimental design, and data analysis. These systems have the potential to significantly accelerate scientific research, reduce costs, and expand access to advanced tools across various fields, including chemistry, biology, and materials science. Recent efforts have demonstrated the potential of LLM-driven agents in supporting researchers with tasks such as literature reviews, experimentation, and report writing. Prominent frameworks, including LitSearch (Ajith et al., 2024), ResearchArena (Kang & Xiong, 2024), SciLitLLM (Li et al., 2024c), CiteME (Press et al., 2024), ResearchAgent (Baek et al., 2024) and Agent Laboratory (Schmidgall et al., 2025), have made strides in automating general research workflows, such as citation management, document discovery, and academic survey generation. However, these systems often lack the domain-specific focus and compliance-driven rigor essential for fields like biomedical domain, where the structured assessment of literature is critical for evidence synthesis.

In tasks like summarization and open-book question answering (QA), Large Language Models (LLMs) often encounter "contextual hallucination", where they produce irrelevant or incorrect responses despite having access to accurate source information. This typically occurs because these models tend to prioritize self-generated content over the input context, causing them to disregard pertinent details. To address this challenge, we introduce a novel method called "Guided Attention Map Editing" (GAME), which dynamically adjusts attention maps to improve contextual relevance. During inference, GAME employs a trained classifier to identify attention maps prone to inducing hallucinations and executes targeted interventions. These interventions, guided by gradient-informed "edit directions'', strategically redistribute attention weights across various heads to effectively reduce hallucination. Comprehensive evaluations on challenging summarization and open-book QA tasks show that GAME consistently reduces hallucinations across a variety of open-source models. Specifically, GAME reduces hallucinations by 10% in the XSum summarization task while achieving a 7X speed-up in computational efficiency compared to the state-of-the-art baselines.

Large language models (LLMs) have demonstrated remarkable capabilities across various tasks, but ensuring their safety and alignment with human values remains crucial. Current safety alignment methods, such as supervised fine-tuning and reinforcement learning-based approaches, can exhibit vulnerabilities to adversarial attacks and often result in shallow safety alignment, primarily focusing on preventing harmful content in the initial tokens of the generated output. While methods like resetting can help recover from unsafe generations by discarding previous tokens and restarting the generation process, they are not well-suited for addressing nuanced safety violations like toxicity that may arise within otherwise benign and lengthy generations. In this paper, we propose a novel backtracking method designed to address these limitations. Our method allows the model to revert to a safer generation state, not necessarily at the beginning, when safety violations occur during generation. This approach enables targeted correction of problematic segments without discarding the entire generated text, thereby preserving efficiency. We demonstrate that our method dramatically reduces toxicity appearing through the generation process with minimal impact to efficiency.

Large Language Models (LLMs), despite their impressive capabilities, often fail to accurately repeat a single word when prompted to, and instead output unrelated text. This unexplained failure mode represents a vulnerability, allowing even end-users to diverge models away from their intended behavior. We aim to explain the causes for this phenomenon and link it to the concept of ``attention sinks'', an emergent LLM behavior crucial for fluency, in which the initial token receives disproportionately high attention scores. Our investigation identifies the neural circuit responsible for attention sinks and shows how long repetitions disrupt this circuit. We extend this finding to other non-repeating sequences that exhibit similar circuit disruptions. To address this, we propose a targeted patch that effectively resolves the issue without negatively impacting the model's overall performance. This study provides a mechanistic explanation for an LLM vulnerability, demonstrating how interpretability can diagnose and address issues, and offering insights that pave the way for more secure and reliable models.

Vision-language models (VLMs) such as CLIP demonstrate strong performance but struggle when adapted to downstream tasks. Prompt learning has emerged as an efficient and effective strategy to adapt VLMs while preserving their pre-trained knowledge. However, existing methods still lead to overfitting and degrade zero-shot generalization. To address this challenge, we propose an optimal transport (OT)-guided prompt learning framework that mitigates forgetting by preserving the structural consistency of feature distributions between pre-trained and fine-tuned models. Unlike conventional point-wise constraints, OT naturally captures cross-instance relationships and expands the feasible parameter space for prompt tuning, allowing a better trade-off between adaptation and generalization. Our approach enforces joint constraints on both vision and text representations, ensuring a holistic feature alignment. Extensive experiments on benchmark datasets demonstrate that our simple yet effective method can outperform existing prompt learning strategies in base-to-novel generalization, cross-dataset evaluation, and domain generalization without additional augmentation or ensemble techniques. The code is available at https://github.com/ChongQingNoSubway/Prompt-OT

An ideal model evaluation should achieve two goals: identifying where the model fails and providing actionable improvement guidance. Toward these goals for Language Model (LM) evaluations, we formulate the problem of generating a weakness profile, a set of weaknesses expressed in natural language, given an LM's performance on every individual instance in a benchmark. We introduce a suite of quantitative assessments to compare different weakness profiling methods. We also propose a weakness profiling method EvalTree. It constructs a capability tree where each node represents a capability described in natural language and is linked to a subset of benchmark instances that specifically evaluate this capability; it then extracts nodes where the LM performs poorly to generate a weakness profile. On the MATH and WildChat benchmarks, we show that EvalTree outperforms baseline weakness profiling methods by identifying weaknesses more precisely and comprehensively. Weakness profiling further enables weakness-guided data collection, and training data collection guided by EvalTree-identified weaknesses improves LM performance more than other data collection strategies. We also show how EvalTree exposes flaws in Chatbot Arena's human-voter-based evaluation practice. To facilitate future work, we release our code and an interface that allows practitioners to interactively explore the capability trees built by EvalTree.