Code Generation & Program Synthesis - Weekly Roundup

A specialized CAD harness drastically improves LLM agent performance, cutting output tokens by up to 88% and boosting F1 scores by 13% through geometry-grounded RL.

Synthesizing data environments from explicit world specifications, not just files, unlocks scalable training for data agents and reveals a fundamental computation-insight gap in frontier models.

Frontier generalist computer-use agents, previously closed-source, are now open for systematic study and community development, with the first release achieving significant performance gains.

Uncovering crucial supporting lemmas for theorem proving by traversing multi-hop dependencies in a formal library's declaration graph.

Autonomous ML engineering agents can achieve frontier-model performance by learning from long experimental trajectories using outcome-guided supervision and windowed reinforcement learning.

A specialized CAD harness drastically improves LLM agent performance, cutting output tokens by up to 88% and boosting F1 scores by 13% through geometry-grounded RL.

Synthesizing data environments from explicit world specifications, not just files, unlocks scalable training for data agents and reveals a fundamental computation-insight gap in frontier models.

Frontier generalist computer-use agents, previously closed-source, are now open for systematic study and community development, with the first release achieving significant performance gains.

Uncovering crucial supporting lemmas for theorem proving by traversing multi-hop dependencies in a formal library's declaration graph.

Autonomous ML engineering agents can achieve frontier-model performance by learning from long experimental trajectories using outcome-guided supervision and windowed reinforcement learning.

Native desktop agents can finally bridge the gap between visual GUI interaction and programmatic code execution by learning a unified action space, achieving state-of-the-art performance on complex real-world tasks.

A neural evaluator predicts Lean 4 proof acceptance with 87.8% accuracy and achieves 4.2x the throughput of a multi-worker Lean baseline, drastically accelerating proof discovery.

Even the best AI agents fail to deliver substantial business value in realistic customer delivery scenarios, struggling to uncover unstated constraints and achieve high impact.

Even the best LLM agents fail nearly half of long-horizon coding tasks due to harness engineering flaws, with failures concentrating on how cross-cutting functionality is modularized.

Open multimodal code intelligence models now rival proprietary systems, driven by a novel rubric-based data synthesis pipeline that precisely evaluates program correctness, code quality, and visual output.

Exhaustive canonical skeletons for symbolic regression reveal that current state-of-the-art methods struggle on a benchmark designed to eliminate structural redundancy and simplicity bias.

Terminal agents can learn to overcome their own mistakes, achieving a 14.6-point Pass@1 improvement on complex coding benchmarks by recovering from failures.

State-of-the-art coding agents exhibit critical boundary violations, with nearly all misinterpreting dependencies and over half compromising environment integrity or user inputs.

Aesthetic code generation is unlocked by using a VLM judge to translate visual design principles from a reference image into actionable rewards for code models.

Fine-tune robot simulation physics with unprecedented interactive control, enabling AI agents to learn from real-world interactions and achieve accurate real-to-sim transfer.

AI-generated mathematical proofs can now be verified at the step-level without human annotation, thanks to a novel framework that leverages formal proof assistants.

Synthesized interactive work environments enable LLMs to achieve state-of-the-art performance on complex agentic benchmarks, outperforming larger models with a 27B parameter model.