Papers for

automation software developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Vision language action models show varied results across household tasks

What Actually Matters? A Paired Evaluation of Inference Settings for $π_{0.5}$ on BEHAVIOR-1K

Abstract: Vision-Language-Action (VLA) models offer a promising paradigm for general-purpose household robotics. However, inference and data-collection strategies that have been deployed in smaller-scale settings do not necessarily generalize to large-scale, diverse task suites. To address this gap, we conduct a systematic, paired evaluation of previously proposed inference and data-collection strategies across all 100 tasks. Our evaluation uses a fine-tuned $π_{0.5}$ model based on the 2025 BEHAVIOR Challenge winner, which achieves a Q-score of 0.39 on the public test set. These results show that proposed strategies do not transfer uniformly to new environments and tasks. Project page, videos, and detailed analysis: https://rubenhammele.github.io/behavior2026

Tue 27 OctRoboticsArtificial IntelligenceComputer Vision and Pattern Recognition
The gist
Vision-language-action models aim to help robots perform many different household tasks by understanding images and language together. The authors studied how well different ways of using these models work across 100 tasks. They found that methods that do well in small experiments don't always work well when scaled to many and varied tasks. The findings come from testing a model tuned for a major robotics challenge, showing that there’s no one-size-fits-all approach yet.
Open → 2610.37917v1

Multi-agent system improves model training with shared learning

Gödel Forest: Balancing Search Depth and Breadth for Data-Centric Recursive Self-Improvement

Abstract: Recursive self-improvement (RSI) aims to achieve compounding gains by having models improve themselves. While most existing RSI systems optimize external agent harnesses or prompts around a frozen base model, data-centric RSI directly updates the model's own parameters by training on agent-generated data. However, because validating data strategies requires expensive model training, existing methods face a fundamental dilemma: a single agent gets trapped in narrow directions and lacks exploration breadth, while naive parallel search or heavy trace sharing sacrifices long-horizon search depth. To address this challenge, we introduce G"odel Forest, a multi-agent framework that organizes recursive self-improvement as an ensemble of co-evolving search trees. In G"odel Forest, each agent autonomously grows a persistent tree, deepening, branching, or pruning data strategies based on model feedback to secure depth, while parallel trees explore distinct regions of the data space to expand breadth. Crucially, rather than leaving trees isolated or flooding them with heavy execution logs, a dynamically co-evolving memory connects the forest: agents continuously distill their successes and failures into compact procedural lessons anchored to a global leaderboard. Through this forest ecosystem, a dead-end in one tree instantly warns the whole forest against unpromising paths, while an empirical breakthrough quickly seeds new exploration branches in neighboring trees. Evaluated on RSIBench-Data across six diverse domains, G"odel Forest outperforms the single-agent baseline by an average of 10.70% while reducing wall-clock time on five tasks. Ablations confirm that co-evolving shared memory yields a +7.00% gain over independent parallel search, demonstrating that collective distillation is key to scalable self-improvement. The code is available at https://github.com/evolvent-ai/Godel-Forest.

Tue 29 SeptComputation and Language
The gist
Improving computer models by teaching themselves can be tricky because they either focus too much on a few ideas or try too many without going deep enough. The authors created G"odel Forest, where many agents work like a team, exploring different ideas and sharing what they learn in a smart way. This teamwork helps models improve faster and avoid wasting time on bad paths. Tests show this method works better and quicker than having just one agent working alone.
Open → 2609.36675v1

Recursive method improves embodied code policies with verified local fixes

RE-0: Verified Recursive Improvement of Embodied Code-as-Policy Agents through Local On-Policy Distillation

Abstract: Code-as-Policy agents accomplish long-horizon embodied tasks by generating and executing code, yet continually improving them with teachers that are stronger but not globally reliable remains a key challenge. Existing distillation methods typically treat the teacher's complete behavior as the supervision target and thus misassign training credit on states where the teacher fails. We propose RE-0, a recursively verified policy improvement framework: rather than assuming that the teacher globally outperforms the student, RE-0 requests local corrections from the teacher on the student's own failure histories and checks in the environment whether each correction is genuinely beneficial; verified corrections yield immediate improvement. Building on this, we propose RE-OPD, which turns verified interventions into supervision for on-policy distillation. Only counterfactually verified teacher interventions provide distribution-level supervision, weighted by their measured local benefit, and the improvement they induce is projected back into the standalone student, so both where supervision is applied and how much credit the teacher receives co-evolve with the student policy. We further prove that the student's per-round gain is lower-bounded by its verified intervention gain up to verification and projection error terms. Experiments across multiple Code-as-Policy embodied tasks show that RE-0 improves both teacher-assisted execution and the standalone student, and generalizes to novel robots and scenes.

Sat 26 SeptRoboticsArtificial IntelligenceMachine Learning
The gist
Some computer programs help robots by writing and running code to complete complicated tasks. But making these programs better is hard when their teachers aren't always right. The authors present a way called RE-0 that asks the teacher only for specific help when the robot itself struggles and then checks if the help actually works. This makes the robot improve more reliably by learning only from useful corrections. Their approach also proved effective in tests with different robots and tasks.
Open → 2609.32416v1

SkillRefine improves industrial planning with better multi-table coordination

SkillRefine: Cross-Source Skill Induction and Execution Validation for LLM Agents in Refinery Planning Software

Abstract: Operating industrial planning software such as AspenTech PIMS (Process Industry Modeling System) requires an LLM agent to combine structural knowledge, procedural knowledge from expert records, and constraints revealed only during execution. These evidence sources are heterogeneous and individually incomplete: documentation describes tables and interfaces but omits task-level coordination, expert CASE records expose multi-table modification patterns without explicit schema grounding, and execution feedback reveals latent constraints only when a plan is executed. We present \textsc{SkillRefine}, a framework that exploits the Documentation--Practice Gap between documented structure and expert practice to extract candidate coordination patterns, ground them against table definitions and COM specifications, and compile them into progressively disclosed skill packages with provenance. The library is then refined through label-free compliance screening, oracle-based match decomposition over table, row, column, and value dimensions, and signal-conditioned trajectory attribution for localized repair. We evaluate \textsc{SkillRefine} on PIMS-Bench, a benchmark built from two AspenTech PIMS demonstration models, using disjoint construction and held-out test tasks. On the held-out test set, \textsc{SkillRefine} achieves absolute component match F1 gains of 14\%--30\% across four LLM backbones, with the largest gains on complex multi-table coordination tasks.

Fri 25 SeptComputational Engineering, Finance, and Science
The gist
Big industrial software systems are complex and hard to operate because information is scattered and incomplete. The authors created SkillRefine, a new method that helps AI agents learn the right step-by-step skills by combining software manuals, expert actions, and feedback after running plans. Their method checks and fixes these skills through several tests without needing extra labels. SkillRefine showed better accuracy, especially when many tables had to work together, helping AI perform refinery planning tasks more reliably.
Open → 2609.30674v1

Agentic AI improves planning for robotic 3D printing with six-axis arms

Kinematics-Grounded Agentic AI for Robotic Additive Manufacturing Process Planning

Abstract: Robotic additive manufacturing (AM) extends material-extrusion printing beyond gantry kinematics but makes process planning robot-dependent. A slicer-generated plan that appears favorable in part coordinates can become infeasible or robotically unfavorable on a manipulator because slicer-process decisions and part orientation determine the generated path, while part orientation and workspace placement affect its kinematic realization. Existing AM tools, large language model (LLM)-based decision-support methods, and digital-shadow systems do not provide integrated pre-execution evaluation of these coupled decisions. This paper presents agentic robotic additive manufacturing (A-RAM), an agent-specialist-tool framework that converts user intent and a part file into traceable, execution-ready plans. The LLM interprets manufacturing objectives and constraints, identifies prescribed and searchable planning variables, and encodes this reasoning in a schema-constrained request; a deterministic Planning Agent instantiates the corresponding search workflow, while domain tools compute quantitative evidence for slicing, placement, inverse kinematics, trajectory timing, Joint-6 jerk, and extrusion. The framework is evaluated on a six-axis robotic-arm AM cell through three case studies covering expert-specified planning, goal-only planning, objective-dependent infill screening, and geometry-dependent orientation-placement selection. Across the evaluated candidate sets, selected plans achieve up to 53.5% lower maximum Joint-6 jerk and 48.3% lower mean absolute Joint-6 jerk than the least favorable valid candidates, while objective-specific infill screening yields motion-plan completion times up to 40.1% shorter and extrusion paths up to 12.7% shorter than the corresponding least favorable screened patterns.

Wed 16 SeptRoboticsArtificial Intelligence
The gist
Robotic 3D printing can be tricky because plans that look good on a computer might not actually work well with a robot’s joint movements and arm positions. The authors created an AI framework that thinks carefully about the robot’s arm and movements before printing. Their system uses AI to understand the printing goals, then uses specialized tools to make and check print plans that work smoothly with the robot’s movements. Their tests showed the plans could greatly reduce robot joint stress and speed up printing compared to less optimized plans.
Open → 2609.19347v1

Benchmark assesses robot learning manipulation skills from videos

Monkey See, Can Monkey Do? A Benchmark for Evaluating Robot Skill Learning by Observation

Abstract: Learning from Observation (LfO) is a fundamental robotic capability that replicates how humans and animals socially learn from each other. Beyond its biological parallels, this modality provides a practical solution for data scaling in sample-inefficient and data-starved domains like robotics. Recent work has demonstrated promising results in learning manipulation skills from human videos, yet progress in this area remains difficult to assess. Existing methods vary widely in assumptions, hardware choices, and environment setups making it difficult to draw meaningful comparisons and identify advances in the field. To address these challenges, we introduce RoboReel: a unified benchmark for evaluating models that learn policies from human videos. RoboReel consists of bundled real-world human demonstration videos, simulated robot trajectories, and evaluation environments on ten manipulation tasks. We develop four test suites to evaluate the models' performance on multiple axes, including the robustness to visual distractors and the ability to complete long-horizon tasks. Our benchmark covers learning-from-observation models from different categories, and studies the effectiveness of multiple representation choices in our benchmark evaluation that covers over seven state-of-the-art algorithms (including our VLA based variants) in the field of LfO. Finally, we present an analysis of the different types of algorithms showing that long-horizon tasks and tasks with low tolerances are still challenging for current models. Webpage: https://roboreel.github.io

Tue 8 SeptRobotics
The gist
Robots can learn how to do tasks by watching videos of humans, similar to how people learn from each other. But it has been hard to compare different approaches because they use different setups and assumptions. The authors created RoboReel, a standard test set with videos, robot actions, and tasks to fairly compare methods for robot learning from human demonstrations. They tested several state-of-the-art algorithms and found that current models still struggle with complex tasks that require many steps or precise accuracy.
Open → 2609.08209v1