AutoAssumption: Searching over Assumptions Discovers ML-Systems Algorithms
cs.AI, cs.LG
Submitted: 2026-11-10
Updated: 2026-11-10
Code: https://github.com/PatrickMassot/leanbluep
Terminology
Sources
- AdaEDL: Early Draft Stopping for Speculative Decoding of Large Language Models via an Entropy-based Lower Bound on Token Acceptance Probability
- RestoreKV: Recovering Full-Cache Behavior Under Aggressive Query-Agnostic KV Cache Eviction
- Verified Code Transpilation with LLMs
- Z-Pruner: Post-Training Pruning of Large Language Models for Efficiency without Retraining
- PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
- Accelerating Large Language Model Decoding with Speculative Sampling
- Let the Barbarians In: How AI Can Accelerate Systems Performance Research
- Barbarians at the Gate: How AI is Upending Systems Research
- Beyond Size: How Gradients Shape Pruning Decisions in Large Language Models
- A Simple and Effective $L_2$ Norm-Based Strategy for KV Cache Compression
- Expected Attention: KV Cache Compression by Estimating Attention from Future Queries Distribution
- Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference
- Simple Baselines are Competitive with Code Evolution
- The Llama 3 Herd of Models
- Automated Design of Agentic Systems
- AIDE: AI-Driven Exploration in the Space of Code
- A Domain-Specific Harness for End-to-End Automation of Optimization Research
- KVzip: Query-Agnostic KV Cache Compression with Context Reconstruction
- Early Discoveries of Algorithmist I: Promise of Provable Algorithm Synthesis at Scale
- ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents