12 citations · 101 across the 80 of their papers we have counts for
86 papers
LoOPSD: On-Policy Self-Distillation with Looped Teachers
Wei Fan, Qingyue Yang, Chao-Hong Tan +6
On-policy self-distillation improves a model using guidance from the model itself, but it still requires a teacher that provides more useful predictions than the student. Existing…
Understanding and Mitigating Hallucination Escape in Tool-Using LLM Agents
Peigui Qi, Kunsheng Tang, Yide Song +2
Large language models (LLMs) increasingly serve as autonomous agents that invoke external tools. However, this capability introduces tool hallucination, selecting incorrect tools o…
GlitchPatch: Repairing Glitch Tokens in Frozen Language Models via Local Retokenization
Kunsheng Tang, Peigui Qi, Yide Song +3
Glitch tokens are anomalous vocabulary entries that can cause large language models (LLMs) to produce outputs inconsistent with their inputs. Existing repair methods require access…
The Safeguard Worked. Is the LLM System Safer?
Pingyu Wu, Weiming Zhang, Nenghai Yu
Safeguards in deployed LLM services are evaluated by refusal, attack success, and policy violation rates. Those rates characterize how a control performed on the requests it was te…
PURA: Provably Unbiased and Robust Multi-Bit Watermarking for AI-Generated Text Attribution
Yaofei Wang, Jinyang Guo, Shuchao Du +6
Fine-grained attribution of AI-generated text is becoming increasingly important for accountability and auditing, yet existing multi-bit watermarking methods still struggle to simu…
Breadcrumbing Search Agents
Xuebin Li, Hanqing Zhao, Siyuan Liang +4
LLM-based search agents are widely used for information-seeking tasks, but their reliance on external tool returns introduces a critical security risk: web content retrieved during…
Safeguards Based on Copyable Context Cannot Provide Reliable Safety for LLMs
Pingyu Wu, Lingyao Zhu, Weiming Zhang +1
Large language model safeguards decide whether to answer before seeing how an answer will be used. This creates a basic problem for dual-use tasks: the same answer can help an auth…
FARI: Robust One-Step Inversion for Watermarking in Diffusion Models
Jindong Yang, Han Fang, Weiming Zhang +2
Inversion-based watermarking is a promising approach to authenticate diffusion-generated images, yet practical use is bottlenecked by inversion that is both slow and error-prone. W…
DNA: Dual-stage Native Attribution for Generated Image Source Tracing
Chao Wang, Kejiang Chen, Zijin Yang +4
The rapid evolution of image generation has produced numerous within-family variants, making source-model attribution of suspect images increasingly important for digital forensics…
BareWave: Waveform-Native Flow-Matching Text-to-Speech
Wei Fan, Chao-Hong Tan, Qian Chen +5
Removing intermediate representations and separately trained decoding stages has become an important direction in generative modeling. In text-to-speech, however, high-quality syst…
TextFake: Benchmarking AI-Generated Image Detection on Text-Rich Images
Yuning Zhang, Changtao Miao, Mingyu Liao +5
Recent AI-generated image (AIGI) detectors perform well on natural-image benchmarks, but their behavior on text-rich forgeries, such as fabricated screenshots, documents, and news…
Provably Secure Agent Guardrail
Benlong Wu, Weiming Zhang, Kejiang Chen +2
As large language models transition from bounded generative engines to agents with expansive execution privileges, AI going out of control precipitates a fundamental crisis in arti…
VLMShield: Efficient and Robust Defense of Vision-Language Models against Malicious Prompts
Peigui Qi, Kunsheng Tang, Yanpu Yu +7
Vision-Language Models (VLMs) face significant safety vulnerabilities from malicious prompt attacks due to weakened alignment during visual integration. Existing defenses suffer fr…
State-Dependent Safety Failures in Multi-Turn Language Model Interaction
Pengcheng Li, Jie Zhang, Tianwei Zhang +5
Safety alignment in large language models is typically evaluated under isolated queries, yet real-world use is inherently multi-turn. Although multi-turn jailbreaks are empirically…
WMVLM: Evaluating Diffusion Model Image Watermarking via Vision-Language Models
Zijin Yang, Yu Sun, Kejiang Chen +4
Digital watermarking is essential for securing generated images from diffusion models. Accurate watermark evaluation is critical for algorithm development, yet existing methods hav…
Character as a Latent Variable in Large Language Models: A Mechanistic Account of Emergent Misalignment and Conditional Safety Failures
Yanghao Su, Wenbo Zhou, Tianwei Zhang +4
Emergent Misalignment refers to a failure mode in which fine-tuning large language models (LLMs) on narrowly scoped data induces broadly misaligned behavior. Prior explanations mai…
SemBind: Binding Diffusion Watermarks to Semantics Against Black-Box Forgery Attacks
Xin Zhang, Zijin Yang, Kejiang Chen +3
Latent-based watermarks, integrated into the generation process of latent diffusion models (LDMs), simplify detection and attribution of generated images. However, recent black-box…
STEAD: Robust Provably Secure Linguistic Steganography with Diffusion Language Model
Yuang Qi, Na Zhao, Qiyi Yao +4
Recent provably secure linguistic steganography (PSLS) methods rely on mainstream autoregressive language models (ARMs) to address historically challenging tasks, that is, to disgu…
Cross-modal Retrieval Models for Stripped Binary Analysis
Guoqiang Chen, Lingyun Ying, Ziyang Song +7
Retrieving binary code via natural language queries is a pivotal capability for downstream tasks in the software security domain, such as vulnerability detection and malware analys…
LiteUpdate: A Lightweight Framework for Updating AI-Generated Image Detectors
Jiajie Lu, Zhenkan Fu, Na Zhao +4
The rapid progress of generative AI has led to the emergence of new generative models, while existing detection methods struggle to keep pace, resulting in significant degradation…
A high-capacity linguistic steganography based on entropy-driven rank-token mapping
Jun Jiang, Weiming Zhang, Nenghai Yu +1
Linguistic steganography enables covert communication through embedding secret messages into innocuous texts; however, current methods face critical limitations in payload capacity…
T2SMark: Balancing Robustness and Diversity in Noise-as-Watermark for Diffusion Models
Jindong Yang, Han Fang, Weiming Zhang +2
Diffusion models have advanced rapidly in recent years, producing high-fidelity images while raising concerns about intellectual property protection and the misuse of generative AI…
EditMark: Watermarking Large Language Models based on Model Editing
Shuai Li, Kejiang Chen, Jun Jiang +5
Large Language Models (LLMs) have demonstrated remarkable capabilities, but their training requires extensive data and computational resources, rendering them valuable digital asse…
SafeGuider: Robust and Practical Content Safety Control for Text-to-Image Models
Peigui Qi, Kunsheng Tang, Wenbo Zhou +5
Text-to-image models have shown remarkable capabilities in generating high-quality images from natural language descriptions. However, these models are highly vulnerable to adversa…