A novel threat is unveiled in which attackers steer the RAG system's response by injecting malicious passages into its knowledge base, enabling the attacker to steer the response without altering the user input or modifying the RAG weights.
Jiaqi Xue, Meng Zheng, Yebowen Hu et al.· arXiv.org· 109 citations· ⚡8
Both the quality loss from reuse and which repairs help depend on the LLM, even between two 8B models are introduced, as well as a common interface for adding new methods and an interactive leaderboard.
This work presents AgentServeSim, a simulator whose unit of execution is the agent program, and proposes LLM-driven automated agent-serving policy search using AgentServeSim as a CPU-based fitness evaluator.
Rakibul Hasan Rajib, Meng Zheng, Qian Lou· 1 citation
Gated-Memory Routing is proposed, which conditions each decision on the query and a learned execution memory, and attains the best average accuracy, exceeding the strongest baseline by 2.44 points, while reducing HumanEval inference cost by 31.9% relative to that baseline.
Rakibul Hasan Rajib, Meng Zheng, Qian Lou· 1 citation
CutBackdoor is presented, the first parameter-supply-chain backdoor that uses cut circuit execution from CutQC as the deployment-time trigger against VQAs, and poisoned parameters preserve full-circuit validation performance while substantially increasing cut-path reconstruction error under noisy finite-shot circuit-cu...
Ahatesham Bhuiyan, Hoang M. Ngo, Cheng Chu et al.· arXiv.org· 0 citations
LAMaS is a latency-aware orchestration framework for learning-based multi-agent systems that achieves the best latency among evaluated learning-based MAS baselines, reducing end-to-end latency by over 50% while maintaining competitive or better accuracy.
Xi Shi, Meng Zheng, Qian Lou· arXiv.org· 2 citations· ⚡1
This work shows that a malicious developer can pair a benign-looking wrapper with crafted metadata to deterministically alter post-generation behavior without modifying model weights, training data, or inference backend, and introduces TIF-BAH, a lightweight middleware defense that verifies wrapper integrity and record...
Nokimul Hasan Arif, Qian Lou, Meng Zheng· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
