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.
hub Canonical reference
AdaEvolve: Adaptive LLM driven zeroth-order optimization
Canonical reference. 80% of citing Pith papers cite this work as background.
26
Pith papers citing it
Background
80% of classified citations
hub tools
citation-role summary
background 4
dataset 1
citation-polarity summary
years
2026 26representative citing papers
FAPO automates LLM pipeline optimization via iterative diagnosis and prompt-or-structure edits, beating GEPA baseline by +14.1 pp mean across 18 comparisons and +33.8 pp when structural changes occur.
MotionDisco discovers long-horizon humanoid loco-manipulation motions from scratch via LLM-guided evolutionary search, trajectory optimization, and pruning, then transfers them to real robots with RL policies.
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
SMCEvolve applies Sequential Monte Carlo sampling to LLM program search with adaptive resampling, mutation mixtures, and convergence control, delivering finite-sample complexity bounds and benchmark gains over prior systems.
FrontierSmith automates synthesis of open-ended coding problems from closed-ended seeds and shows measurable gains on two open-ended LLM coding benchmarks.
DeInfer reduces parallel inference communication cost for decomposed LLMs by up to 78% by moving collective operations into the low-rank latent space and redesigning KV-cache reconstruction for static graph compatibility.
Meta-Harness discovers improved harness code for LLMs via agentic search over prior execution traces, yielding 7.7-point gains on text classification with 4x fewer tokens and 4.7-point gains on math reasoning across held-out models.
Different patterns of environmental change (space vs time) require different LLM update behaviors; no single family of methods—prompts, distillation, RL, or compression—handles all regimes.
Evolution Fine-Tuning trains LLMs on 156K trajectories spanning 371 tasks to achieve 10.22% average improvement on 22 held-out optimization tasks and match SOTA on select circle-packing problems when combined with test-time RL.
CHIA is a distributed graph-based framework for composing, deploying, and verifying agentic AI hardware/software co-design flows, demonstrated on RTL, simulation, and compiler tasks.
An agent harness combining staged task decomposition, multimodal evidence tooling, and artifact-grounded self-improvement scores 81.0 GRAS on multimodal scientific curation, 22.4 points above the strongest baseline — with the caveat that 8 of 23 evaluation papers were used for optimization.
AIChilles finds 49 distinct hidden weaknesses across 30 AI-evolved programs in five applications by combining workload extraction, agent-based constraint inference, differential oracles, and coverage to expose regressions.
Evoflux applies evolutionary search at inference time to repair executable tool workflows for compact agents, outperforming SFT and SFT+DPO on held-out MCP-Bench tasks with live servers and 250 tools.
BenchEvolver evolves coding problem solutions to generate harder, valid tasks, producing LiveCodeBench-Plus where frontier models score 27.5-62.6% and enabling RL gains on held-out tests.
VESTA introduces dynamic tool creation for VLMs that outperforms static-tool and no-tool baselines on distribution fitting, time series, and astronomy tasks in the new DAWN benchmark.
ScientistOne introduces Chain-of-Evidence and an audit system that achieves zero hallucinated references, perfect score verification, and top method-code alignment while matching or beating human experts on five frontier tasks and generalizing to six more.
An LLM-driven dual-loop controller claims 72.5% stable-phase improvement over default and 153.3% over the strongest non-LLM baseline, but the comparison protocol inflates the gaps by scoring baselines during continued exploration.
A dual-population evolutionary ensemble of coding agents discovers a rescale-then-interpolate PE for ICON example-count generalization and outperforms static-agent baselines via stage-dependent adaptation.
CTM-AI combines a formal consciousness model with foundation models to report state-of-the-art results on sarcasm detection, humor, and agentic tool-use benchmarks.
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.
SimpleTES scales test-time evaluation in LLMs to discover state-of-the-art solutions on 21 scientific problems across six domains, outperforming frontier models and optimization pipelines with examples like 2x faster LASSO and new Erdos constructions.
EurekAgent achieves new state-of-the-art results on mathematics, kernel engineering, and machine learning tasks by engineering agent environments for autonomous scientific discovery, including a 26-circle packing result at under $11 API cost.
EEVEE introduces a router-based multi-dataset test-time prompt learning framework for LLM agents that uses router-prompt co-evolution to improve robustness on heterogeneous data streams.
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.
-
FAPO: Fully Automated Prompt Optimization of Multi-Step LLM Pipelines
FAPO automates LLM pipeline optimization via iterative diagnosis and prompt-or-structure edits, beating GEPA baseline by +14.1 pp mean across 18 comparisons and +33.8 pp when structural changes occur.
-
MotionDisco: Motion Discovery for Extreme Humanoid Loco-Manipulation
MotionDisco discovers long-horizon humanoid loco-manipulation motions from scratch via LLM-guided evolutionary search, trajectory optimization, and pruning, then transfers them to real robots with RL policies.
-
What Do Evolutionary Coding Agents Evolve?
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
-
SMCEvolve: Principled Scientific Discovery via Sequential Monte Carlo Evolution
SMCEvolve applies Sequential Monte Carlo sampling to LLM program search with adaptive resampling, mutation mixtures, and convergence control, delivering finite-sample complexity bounds and benchmark gains over prior systems.
-
FrontierSmith: Synthesizing Open-Ended Coding Problems at Scale
FrontierSmith automates synthesis of open-ended coding problems from closed-ended seeds and shows measurable gains on two open-ended LLM coding benchmarks.
-
Co-evolving Agent Architectures and Interpretable Reasoning for Automated Optimization
DeInfer reduces parallel inference communication cost for decomposed LLMs by up to 78% by moving collective operations into the low-rank latent space and redesigning KV-cache reconstruction for static graph compatibility.
-
Meta-Harness: End-to-End Optimization of Model Harnesses
Meta-Harness discovers improved harness code for LLMs via agentic search over prior execution traces, yielding 7.7-point gains on text classification with 4x fewer tokens and 4.7-point gains on math reasoning across held-out models.
-
When Does Continual Learning Require Learning
Different patterns of environmental change (space vs time) require different LLM update behaviors; no single family of methods—prompts, distillation, RL, or compression—handles all regimes.
-
Evolution Fine-Tuning: Learning to Discover Across 371 Optimization Tasks
Evolution Fine-Tuning trains LLMs on 156K trajectories spanning 371 tasks to achieve 10.22% average improvement on 22 held-out optimization tasks and match SOTA on select circle-packing problems when combined with test-time RL.
-
CHIA: An open-source framework for principled, agentic AI-driven hardware/software co-design research
CHIA is a distributed graph-based framework for composing, deploying, and verifying agentic AI hardware/software co-design flows, demonstrated on RTL, simulation, and compiler tasks.
-
Building Agent Harnesses for Scientific Curation from Multimodal Sources
An agent harness combining staged task decomposition, multimodal evidence tooling, and artifact-grounded self-improvement scores 81.0 GRAS on multimodal scientific curation, 22.4 points above the strongest baseline — with the caveat that 8 of 23 evaluation papers were used for optimization.
-
AIChilles: Automatically Uncovering Hidden Weaknesses in AI-Evolved Systems
AIChilles finds 49 distinct hidden weaknesses across 30 AI-evolved programs in five applications by combining workload extraction, agent-based constraint inference, differential oracles, and coverage to expose regressions.
-
Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents
Evoflux applies evolutionary search at inference time to repair executable tool workflows for compact agents, outperforming SFT and SFT+DPO on held-out MCP-Bench tasks with live servers and 250 tools.
-
BenchEvolver: Frontier Task Synthesis via Solution-Centric Evolution
BenchEvolver evolves coding problem solutions to generate harder, valid tasks, producing LiveCodeBench-Plus where frontier models score 27.5-62.6% and enabling RL gains on held-out tests.
-
VESTA: Visual Exploration with Statistical Tool Agents
VESTA introduces dynamic tool creation for VLMs that outperforms static-tool and no-tool baselines on distribution fitting, time series, and astronomy tasks in the new DAWN benchmark.
-
ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence
ScientistOne introduces Chain-of-Evidence and an audit system that achieves zero hallucinated references, perfect score verification, and top method-code alignment while matching or beating human experts on five frontier tasks and generalizing to six more.
-
TuxBot: Semantic-Aware Online OS Tuning with Large Language Models
An LLM-driven dual-loop controller claims 72.5% stable-phase improvement over default and 153.3% over the strongest non-LLM baseline, but the comparison protocol inflates the gaps by scoring baselines during continued exploration.
-
Evolutionary Ensemble of Agents
A dual-population evolutionary ensemble of coding agents discovers a rescale-then-interpolate PE for ICON example-count generalization and outperforms static-agent baselines via stage-dependent adaptation.
-
CTM-AI: A Blueprint for General AI Inspired by a Model of Consciousness
CTM-AI combines a formal consciousness model with foundation models to report state-of-the-art results on sarcasm detection, humor, and agentic tool-use benchmarks.
-
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.
-
Evaluation-driven Scaling for Scientific Discovery
SimpleTES scales test-time evaluation in LLMs to discover state-of-the-art solutions on 21 scientific problems across six domains, outperforming frontier models and optimization pipelines with examples like 2x faster LASSO and new Erdos constructions.
-
EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery
EurekAgent achieves new state-of-the-art results on mathematics, kernel engineering, and machine learning tasks by engineering agent environments for autonomous scientific discovery, including a 26-circle packing result at under $11 API cost.
-
EEVEE: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents
EEVEE introduces a router-based multi-dataset test-time prompt learning framework for LLM agents that uses router-prompt co-evolution to improve robustness on heterogeneous data streams.
-
Declarative Data Services: Structured Agentic Discovery for Composing Data Systems
DDS introduces typed contracts at intent, operator DAG, skills, and runtime layers to bound agentic search for data system compositions, achieving convergence on a trading workload where unbounded iteration fails.
-
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.