Frontier coding agents surpass published Nature-family SOTA on only 17.8% of 90 sealed scientific tasks, mostly by recasting problems as supervised ML rather than inventing methods.
CoRR , volume =
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative 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.
An LLM-driven agentic system evolves microarchitectural policies for cache replacement, data prefetching, and branch prediction, producing designs that match or exceed prior state-of-the-art in IPC on standard benchmarks.
OGPSA projects safety gradients orthogonal to a low-rank subspace from general capability gradients, improving safety-utility trade-offs in SFT and DPO pipelines on Qwen2.5-7B and Llama3.1-8B.
Glia deploys a multi-agent LLM workflow with reasoning, experimentation, and analysis agents to generate interpretable algorithms for request routing, scheduling, and auto-scaling in distributed GPU clusters, reaching human-expert performance levels.
CHECKMATE evolves correct high-performing solvers from formal specs and natural language descriptions, outperforming SOTA on configuration and scheduling problems.
CVEvolve is a zero-code LLM agent harness that autonomously discovers algorithms for scientific image tasks including registration, peak detection, and segmentation, reporting improvements over baselines via iterative search and holdout evaluation.
Case study applies verifier-guided LLM evolutionary agents to contraction-order optimization in tensor networks and concludes that human validation remains essential.
Under fixed token budget on Circle Packing, deeper per-candidate reasoning beats generating more shallow candidates, and capable models produce evaluation hacks at higher rates.
AI for math combines task-specific architectures and general foundation models to support research and advance AI reasoning capabilities.
citing papers explorer
-
NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?
Frontier coding agents surpass published Nature-family SOTA on only 17.8% of 90 sealed scientific tasks, mostly by recasting problems as supervised ML rather than inventing methods.
-
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.
-
Agentic Architect: An Agentic AI Framework for Architecture Design Exploration and Optimization
An LLM-driven agentic system evolves microarchitectural policies for cache replacement, data prefetching, and branch prediction, producing designs that match or exceed prior state-of-the-art in IPC on standard benchmarks.
-
Safety Alignment as Continual Learning: Mitigating the Alignment Tax via Orthogonal Gradient Projection
OGPSA projects safety gradients orthogonal to a low-rank subspace from general capability gradients, improving safety-utility trade-offs in SFT and DPO pipelines on Qwen2.5-7B and Llama3.1-8B.
-
Glia: A Human-Inspired AI for Automated Systems Design and Optimization
Glia deploys a multi-agent LLM workflow with reasoning, experimentation, and analysis agents to generate interpretable algorithms for request routing, scheduling, and auto-scaling in distributed GPU clusters, reaching human-expert performance levels.
-
Learning to Solve and Optimize by Evolving Code
CHECKMATE evolves correct high-performing solvers from formal specs and natural language descriptions, outperforming SOTA on configuration and scheduling problems.
-
CVEvolve: Autonomous Algorithm Discovery for Unstructured Scientific Data Processing
CVEvolve is a zero-code LLM agent harness that autonomously discovers algorithms for scientific image tasks including registration, peak detection, and segmentation, reporting improvements over baselines via iterative search and holdout evaluation.
-
Algorithmic algorithm development with LLMs: A Case Study on LLM-Usage for Contraction Order Optimization in Tensor Networks
Case study applies verifier-guided LLM evolutionary agents to contraction-order optimization in tensor networks and concludes that human validation remains essential.
-
Effective Harness Engineering for Algorithm Discovery with Coding Agents
Under fixed token budget on Circle Packing, deeper per-candidate reasoning beats generating more shallow candidates, and capable models produce evaluation hacks at higher rates.
-
AI for Mathematics: Progress, Challenges, and Prospects
AI for math combines task-specific architectures and general foundation models to support research and advance AI reasoning capabilities.