Methods for improving knowledge use in large language models typically fall into two regimes. Non-parametric retrieval offers flexible access to external knowledge, but adds retrieval latency, context overhead, and only shallow integration with the backbone. Parametric adaptation is efficient at inference time, but ent...
Ming-Yuan Li, Guang-Sheng Yu, Xu Wang et al.· 3 citations
K-Bench is introduced, a benchmark that scores LLM unlearning under agentic deployment and certifies forgetting by reading the model's final answer, where a model that refuses to answer already counts as having forgotten.
Guang-Sheng Yu, Yan-Na Jiang, Qin Wang et al.· 0 citations
Traffic Signal Control (TSC) is a safety-critical cyber-physical system that relies on real-time sensing. Corrupted observations caused by adversarial perturbations or sensor failures can propagate from the sensing layer into the controller and degrade traffic efficiency. Existing robust Reinforcement Learning (RL)-bas...
Ming-Yuan Li, Chun-Yu Liu, Xiao Liu et al.· 0 citations
A residual-stream covert-channel attack is introduced that maps messages to codewords and injects them into an intermediate residual stream through a compromised runtime hook and is scaled with the local residual norm using the signal-to-residual-norm ratio.
Ming-Yuan Li, Yan-Na Jiang, Guang-Sheng Yu et al.· 0 citations
The results support generated memory as a selective correction to direct retrieval and highlight routing when, which, and how strongly to intervene as the central challenge.
Ming-Yuan Li, Guang-Sheng Yu, Ju-Yuan Zhang et al.· 0 citations
Unlearning benchmarks such as TOFU and MUSE certify forgetting by reading the model's final answer, where a model that refuses to answer already counts as having forgotten. We show that this model-level certificate does not transfer once the model is deployed as an agent. We introduce K-Bench, a benchmark that scores L...
Guangsheng Yu, Yanna Jiang, Qin Wang et al.· 0 citations
The results suggest that Engram can serve as a reusable external knowledge artifact, provided that the target has access to a compatible reader interface and target-side adaptation can further improve alignment when direct reader reuse is insufficient.
Mingyuan Li, Guangsheng Yu, Xu Wang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
