English
Related papers

Related papers: CIVeX: Causal Intervention Verification for Langua…

200 papers

Evaluating whether explanations faithfully reflect a model's reasoning remains an open problem. Existing benchmarks use single interventions without statistical testing, making it impossible to distinguish genuine faithfulness from…

Computation and Language · Computer Science 2026-03-20 Abhinaba Basu , Pavan Chakraborty

Causal discovery is a powerful technique for identifying causal relationships among variables in data. It has been widely used in various applications in software engineering. Causal discovery extensively involves conditional independence…

Software Engineering · Computer Science 2023-09-12 Pingchuan Ma , Zhenlan Ji , Peisen Yao , Shuai Wang , Kui Ren

AI agents that execute tasks via tool calls frequently hallucinate results - fabricating tool executions, misstating output counts, or presenting inferences as facts. Recent approaches to verifiable AI inference rely on zero-knowledge…

Cryptography and Security · Computer Science 2026-03-12 Abhinaba Basu

Large language models (LLMs) remain acutely vulnerable to prompt injection and related jailbreak attacks; heuristic guardrails (rules, filters, LLM judges) are routinely bypassed. We present Contextual Integrity Verification (CIV), an…

Cryptography and Security · Computer Science 2025-08-20 Aayush Gupta

Explanations for AI models in high-stakes domains like medicine often lack verifiability, which can hinder trust. To address this, we propose an interactive agent that produces explanations through an auditable sequence of actions. The…

Artificial Intelligence · Computer Science 2025-11-04 Yuhang Huang , Zekai Lin , Fan Zhong , Lei Liu

Why do language agents fail on tasks they are capable of solving? We argue that many such failures are reliability failures caused by stochastic drift from a task's latent solution structure, not capability failures. Every well-defined…

Computation and Language · Computer Science 2026-02-24 Wilson Y. Lee

Measuring long-run LLM outcomes (user satisfaction, expert judgment, downstream KPIs) is expensive. Teams default to cheap LLM judges, but uncalibrated proxies can invert rankings entirely. Causal Judge Evaluation (CJE) makes it affordable…

Methodology · Statistics 2026-01-22 Eddie Landesberg , Manjari Narayan

Large language model (LLM) agents frequently fail on multi-step tasks involving reasoning, tool use, and environment interaction. While such failures are typically logged or retried heuristically, they contain structured signals about where…

Machine Learning · Computer Science 2026-05-26 Akash Bonagiri , Devang Borkar , Gerard Janno Anderias , Setareh Rafatirad , Houman Homayoun

Vulnerability detection tools are widely adopted in software projects, yet they often overwhelm maintainers with false positives and non-actionable reports. Automated exploitation systems can help validate these reports; however, existing…

Cryptography and Security · Computer Science 2026-02-17 Amirali Sajadi , Tu Nguyen , Kostadin Damevski , Preetha Chatterjee

Recent methods for improving LLM mathematical reasoning, whether through MCTS-based test-time search or causal graph-guided knowledge injection, cannot identify which concepts causally contribute to a correct answer, as the observed…

Machine Learning · Computer Science 2026-05-11 Tsuyoshi Okita

Causal inference is to estimate the causal effect in a causal relationship when intervention is applied. Precisely, in a causal model with binary interventions, i.e., control and treatment, the causal effect is simply the difference between…

Machine Learning · Computer Science 2022-07-19 Zhenyu Lu , Yurong Cheng , Mingjun Zhong , George Stoian , Ye Yuan , Guoren Wang

Tool-using LLM agents must act on untrusted webpages, emails, files, and API outputs while issuing privileged tool calls. Existing defenses often mediate trust at the granularity of an entire tool invocation, forcing a brittle choice in…

Cryptography and Security · Computer Science 2026-05-13 Linfeng Fan , Ziwei Li , Yuan Tian , Yichen Wang , Rongsheng Li , Xiong Wang

LLM agents operating in open environments face escalating risks from indirect prompt injection, particularly within the tool stream where manipulated metadata and runtime feedback hijack execution flow. Existing defenses encounter a…

Cryptography and Security · Computer Science 2026-01-15 Junda Lin , Zhaomeng Zhou , Zhi Zheng , Shuochen Liu , Tong Xu , Yong Chen , Enhong Chen

Evaluating LLM-generated interactive software requires execution in addition to static analysis. The key difficulty is that correctness is a graph-level reachable property over latent UI state-transition graphs, whereas a GUI evaluator…

Software Engineering · Computer Science 2026-05-20 Sirui Hong , Zhijie Liu , Tengfei Li , Wei Tao , Yifan Wu , Chenglin Wu

Causal probing aims to analyze foundation models by examining how intervening on their representation of various latent properties impacts their outputs. Recent works have cast doubt on the theoretical basis of several leading causal…

Machine Learning · Computer Science 2025-12-23 Marc Canby , Adam Davies , Chirag Rastogi , Julia Hockenmaier

Agent skills extend LLM agents with privileged third-party capabilities such as filesystem access, credentials, network calls, and shell execution. Existing safety work catches malicious prompts and risky runtime actions, but the skill…

Cryptography and Security · Computer Science 2026-05-13 Yuhao Wu , Tung-Ling Li , Hongliang Liu

Large language model agents are increasingly expected to perform operational work: calling APIs, manipulating files, assembling workflows, and acting inside enterprise systems. Yet the tool layer on which this execution depends is still…

Software Engineering · Computer Science 2026-05-28 Swanand Rao

Machine learning and Large language models (LLMs) for vulnerability detection has received significant attention in recent years. Unfortunately, state-of-the-art techniques show that LLMs are unsuccessful in even distinguishing the…

Cryptography and Security · Computer Science 2025-08-05 Mohammed Sayagh , Mohammad Ghafari

OpenClaw's ClawHub marketplace hosts tens of thousands of community-contributed agent skills (49,592 in our 2026-04-04 snapshot), and recent audits report that 13-26% contain security vulnerabilities. Regex scanners miss obfuscated…

Cryptography and Security · Computer Science 2026-05-27 Yinghan Hou , Zongyou Yang , Zaihu Pang , Xiujun Ma

Agentic vision-language models are increasingly trained to "think with images" by calling image operations. However, we show that high final-answer accuracy often hides unfaithful visual reasoning: models may invoke tools on irrelevant…

Computer Vision and Pattern Recognition · Computer Science 2026-03-03 Xinhai Hou , Shaoyuan Xu , Manan Biyani , Moyan Li , Jia Liu , Todd C. Hollon , Bryan Wang
‹ Prev 1 2 3 10 Next ›