SKIM: Pruning Large Language Model Agents via Selective Knowledge Informed Masking
cs.CL, cs.AI
Submitted: 2026-11-05
Updated: 2026-11-05
Comments: NeurIPS 2026. Project page: https://moon4sake.github.io/skim
Project page: https://moon4sake.github.io/skim
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- JSONSchemaBench: A Rigorous Benchmark of Structured Outputs for Language Models
- Distilling the Knowledge in a Neural Network
- Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programs
- Knowledge Distillation of Transformer-based Language Models Revisited
- GPT-4 Technical Report
- LLM Pruning and Distillation in Practice: The Minitron Approach
- Efficient Guided Generation for Large Language Models
- Qwen3 Technical Report
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data