8 papers
Spherical Steering: Geometry-Aware Activation Rotation for Language Models
Zejia You, Chunyuan Deng, Hanjie Chen
Inference-time steering offers a promising way to control language models (LMs) without retraining. However, standard approaches typically rely on activation addition, which inevit…
The Generalization Ridge: Information Flow in Natural Language Generation
Ruidi Chang, Chunyuan Deng, Hanjie Chen
Transformer-based language models have achieved state-of-the-art performance in natural language generation (NLG), yet their internal mechanisms for synthesizing task-relevant info…
Knowing When to Stop: Efficient Context Processing via Latent Sufficiency Signals
Roy Xie, Junlin Wang, Paul Rosu +4
Large language models (LLMs) process entire input contexts indiscriminately, which is inefficient when the information required to answer a query is localized within the context. W…
Language Models are Symbolic Learners in Arithmetic
Chunyuan Deng, Zhiqi Li, Roy Xie +2
The prevailing question in LM performing arithmetic is whether these models learn to truly compute or if they simply master superficial pattern matching. In this paper, we argues f…
Steering Information Utility in Key-Value Memory for Language Model Post-Training
Chunyuan Deng, Ruidi Chang, Hanjie Chen
Recent advancements in language models (LMs) have marked a shift toward the growing importance of post-training. Yet, post-training approaches such as supervised fine-tuning (SFT)…
Rethinking Diverse Human Preference Learning through Principal Component Analysis
Feng Luo, Rui Yang, Hao Sun +5
Understanding human preferences is crucial for improving foundation models and building personalized AI systems. However, preferences are inherently diverse and complex, making it…