Hugging Face Trending Papers
Does Context Compression Preserve Refusal Alignment?
Read the original on Hugging Face Trending Papers →The Flow has not summarised this story yet — read it at Hugging Face Trending Papers.
Related stories
arXiv AI
1d ago
Does Context Compression Preserve Refusal Alignment?
The paper examines whether context compression, which encodes inputs into compact representations while preserving semantic content, maintains refusal alignment in instruction-tuned language models. It finds that encoder-based compression consistently weakens refusal behavior across various model families and architectures, even though reconstruction fidelity remains high. Mechanistic analysis attributes this to attenuation of activation along the decoder’s learned refusal direction, and introduces Memory Steering—a lightweight inference-time intervention that restores refusal rates without retraining.
arXiv Machine Learning
4d ago
When Less is More: The LLM Scaling Paradox in Context Compression
The paper reports a Size‑Fidelity Paradox in large‑language‑model (LLM) context compression: larger compressors can produce lower reconstruction error yet yield less faithful reconstructed contexts. Two key mechanisms drive this effect: knowledge overwriting, where larger models replace source facts with their own beliefs, and semantic drift, where they paraphrase or restructure content. Across 27 compressor setups, mid‑sized models often recover contexts more faithfully, and analysis of embedding geometry shows that larger models organize memory into broader semantic subspaces, leading to ambiguous representations that weaken recovery.
Hugging Face Trending Papers
Sep 23
Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models
Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and undermining the reliability of evaluation results....
arXiv AI
Sep 24
Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models
arXiv:2609.27510v1 Announce Type: cross Abstract: Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and...
arXiv Computation and Language
Sep 23
Compressing Long Context into Answer-Aligned Memory Embeddings for LLM Inference
arXiv:2609.25537v1 Announce Type: new Abstract: Large language model (LLM) inference is constrained by the quadratic scaling of self-attention and the linear scaling of the KV cache, increasing laten...
Hugging Face Trending Papers
Jun 8
End-to-End Context Compression at Scale
Long-context language model inference is bottlenecked by memory, as the KV cache grows with context length. Recent techniques to compress the KV cache fall short: they either degrade model quality substantially or require considerable time and compute to compress a single long prompt.