7 papers
Diversity Matters: Distributional Feature Coverage Sample Selection for Data-Efficient Backdoor Attacks
Yi Yang, Xiaoke Chen, Jinyang Huang +6
Backdoor attacks compromise training data so that a model retains clean accuracy but predicts an attacker-chosen target on triggered inputs. At very low poisoning rates, only a few…
AcademiClaw: When Students Set Challenges for AI Agents
Junjie Yu, Pengrui Lu, Weiye Si +75
Benchmarks within the OpenClaw ecosystem have thus far evaluated exclusively assistant-level tasks, leaving the academic-level capabilities of OpenClaw largely unexamined. We intro…
Checkerboard: A Simple, Effective, Efficient and Learning-free Clean Label Backdoor Attack with Low Poisoning Budget
Yi Yang, Jinyang Huang, Binbin Liu +5
Backdoor attacks threaten the deep learning supply chain by poisoning a small fraction of the training data so that a model behaves normally on clean inputs but misclassifies trigg…
LBLLM: Lightweight Binarization of Large Language Models via Three-Stage Distillation
Siqing Song, Chuang Wang, Yong Lang +2
Deploying large language models (LLMs) in resource-constrained environments is hindered by heavy computational and memory requirements. We present LBLLM, a lightweight binarization…
TASU2: Controllable CTC Simulation for Alignment and Low-Resource Adaptation of Speech LLMs
Jing Peng, Chenghao Wang, Yi Yang +5
Speech LLM post-training increasingly relies on efficient cross-modal alignment and robust low-resource adaptation, yet collecting large-scale audio-text pairs remains costly. Text…
Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference
Quantong Qiu, Zhiyi Hong, Yi Yang +5
The quadratic computational complexity of standard attention mechanisms presents a severe scalability bottleneck for LLMs in long-context scenarios. While hybrid attention mechanis…