Chua Jia Jim Deryl — accepted AI-conference papers

2026

Fine-Grained Activation Steering: Steering Less, Achieving More

Zijian Feng, Tianjiao Li, Zixiao Zhu, Hanzhang Zhou, Junlang Qian, Li Zhang +4

ICLR 2026poster

Activation steering has emerged as a cost-effective paradigm for modifying large language model (LLM) behaviors. Existing methods typically intervene at the block level, steering the bundled activations of selected attention heads, feedforward networks, or residual streams. However, we reveal that b…

Cited by 0 Source ↗code ↗Scholar ↗

2025

Restoring Pruned Large Language Models via Lost Component Compensation

Zijian Feng, Hanzhang Zhou, Zixiao Zhu, Tianjiao Li, Chua Jia Jim Deryl, Mak Lee Onn +2

NeurIPS 2025spotlight

Pruning is a widely used technique to reduce the size and inference cost of large language models (LLMs), but it often causes performance degradation. To mitigate this, existing restoration methods typically employ parameter-efficient fine-tuning (PEFT), such as LoRA, to recover the pruned model's p…

Cited by 0 Source ↗Scholar ↗