ShapeBench is a new benchmark suite for aerodynamic shape optimization with 103 tasks showing high variance in optimizer rankings across categories.
hub
Mathematical discoveries from program search with large language models
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
A knowledge-first approach to LLM-driven automatic heuristic design in combinatorial optimization yields better discovery efficiency, transfer, and generalization than code-centric baselines by formalizing a distortion-compression trade-off.
LLM-AutoSciLab proposes an LLM-driven closed-loop system for hypothesis generation and adaptive experiment selection that reports higher accuracy and 2-5x better sample efficiency than baselines on new chemistry and gene-network discovery benchmarks.
RMA, a multi-agent system with structured memory and iterative feedback loops, solves 8 out of 10 research-level math problems on the new First Proof benchmark and outperforms GPT-5.2R and Aletheia according to expert evaluation.
DrugSAGE accumulates cross-task memory of skills, statistical evidence, and recurring errors to let LLM agents achieve top-ranked performance on molecular property prediction tasks with reduced or zero test-time search.
ADKO is a decentralized framework where agents share compact GP-derived tokens and LM insights to achieve collaborative Bayesian optimization with a decomposed regret bound that includes compression and approximation losses.
HMACE deploys Proposer, Generator, Evaluator, and Reflector agents in an evolutionary loop to generate and refine heuristics for NP-hard problems, reporting lower optimality gaps and token costs than baselines on TSP and Online BPP.
A eight-agent system automatically reproduces and improves SOTA models from top AI papers, reporting 105 surpassing variants on a curated 125-paper set.
An LLM-driven evolutionary framework that writes and evolves Python-coded trading factors reports large gains in predictive accuracy on CSI300, but with incomplete validation.
LLM-FE is a framework that treats feature engineering as LLM-driven program search with data feedback, reporting consistent gains over baselines on classification and regression tabular tasks.
PACEvolve++ uses a phase-adaptive reinforcement learning advisor to decouple hypothesis selection from execution in LLM-driven evolutionary search, delivering faster convergence than prior frameworks on load balancing, recommendation, and protein tasks.
citing papers explorer
-
ShapeBench: A Scalable Benchmark and Diagnostic Suite for Standardized Evaluation in Aerodynamic Shape Optimization
ShapeBench is a new benchmark suite for aerodynamic shape optimization with 103 tasks showing high variance in optimizer rankings across categories.
-
Back to the Beginning of Heuristic Design: Bridging Code and Knowledge with LLMs
A knowledge-first approach to LLM-driven automatic heuristic design in combinatorial optimization yields better discovery efficiency, transfer, and generalization than code-centric baselines by formalizing a distortion-compression trade-off.
-
LLM-AutoSciLab: Closed-Loop Scientific Discovery via Active Experimentation with LLMs
LLM-AutoSciLab proposes an LLM-driven closed-loop system for hypothesis generation and adaptive experiment selection that reports higher accuracy and 2-5x better sample efficiency than baselines on new chemistry and gene-network discovery benchmarks.
-
RMA: an Agentic System for Research-Level Mathematical Problems
RMA, a multi-agent system with structured memory and iterative feedback loops, solves 8 out of 10 research-level math problems on the new First Proof benchmark and outperforms GPT-5.2R and Aletheia according to expert evaluation.
-
DrugSAGE:Self-evolving Agent Experience for Efficient State-of-the-Art Drug Discovery
DrugSAGE accumulates cross-task memory of skills, statistical evidence, and recurring errors to let LLM agents achieve top-ranked performance on molecular property prediction tasks with reduced or zero test-time search.
-
ADKO: Agentic Decentralized Knowledge Optimization
ADKO is a decentralized framework where agents share compact GP-derived tokens and LM insights to achieve collaborative Bayesian optimization with a decomposed regret bound that includes compression and approximation losses.
-
HMACE: Heterogeneous Multi-Agent Collaborative Evolution for Combinatorial Optimization
HMACE deploys Proposer, Generator, Evaluator, and Reflector agents in an evolutionary loop to generate and refine heuristics for NP-hard problems, reporting lower optimality gaps and token costs than baselines on TSP and Online BPP.
-
AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discovery
A eight-agent system automatically reproduces and improves SOTA models from top AI papers, reporting 105 surpassing variants on a curated 125-paper set.
-
Cognitive Alpha Mining via LLM-Driven Code-Based Evolution
An LLM-driven evolutionary framework that writes and evolves Python-coded trading factors reports large gains in predictive accuracy on CSI300, but with incomplete validation.
-
LLM-FE: Automated Feature Engineering for Tabular Data with LLMs as Evolutionary Optimizers
LLM-FE is a framework that treats feature engineering as LLM-driven program search with data feedback, reporting consistent gains over baselines on classification and regression tabular tasks.
-
PACEvolve++: Improving Test-time Learning for Evolutionary Search Agents
PACEvolve++ uses a phase-adaptive reinforcement learning advisor to decouple hypothesis selection from execution in LLM-driven evolutionary search, delivering faster convergence than prior frameworks on load balancing, recommendation, and protein tasks.