A three-stage pipeline mines skill libraries from GUI interaction data, achieving high purity in five of eight clusters against InteraSkill labels. However, the method only slightly improves skill-step accuracy on IW and fails to advance performance on BrowseComp+ or key metrics, indicating limitations in cross-domain policy transfer.
AutoPass uses runtime and compiler evidence to guide LLM-generated optimization decisions, outperforming expert heuristics and classical autotuning methods. It achieves geometric-mean speedups of 1.043x on x86-64 and 1.117x on ARM64 systems without prior training or fine-tuning.
Tri-Info uses information theory to detect failures in Vision-Language-Action models by analyzing action diversity, temporal consistency, and state coupling. It achieves 83% accuracy on real-world tasks across six models and three environments, outperforming prior methods and maintaining performance without retraining.
A new framework enables large language models to develop 'Connect the Dots' capability, allowing long-lifecycle agents to learn from experiences and iteratively update their environment context. The framework uses reinforcement learning with long rollout sequences and custom tasks to promote cross-domain generalization, showing effective out-of-distribution performance in both domains and transition settings.
A hierarchical system using a pretrained LLM to select RL skill policies outperforms flat RL in a 2v2 King of the Hill environment. It matches hand-crafted behavior tree performance in win rate and is perceived as more human-like by 60% of users, highlighting effective coordination and adaptability without manual rule design.
The AI Economist Agent uses RAG, knowledge graphs, and LLMs to generate economic narratives grounded in theory and data. It enables model-based analysis, evidence retrieval, and report generation, ensuring economic coherence and traceability through explicit model computations.
A sensorimotor world model (SMWM) is introduced that learns compact, action-aligned latent representations from offline trajectories. It uses inverse dynamics regularization to prevent representation collapse and enable stable, interpretable world models without requiring frozen encoders or complex regularizers. SMWM achieves competitive planning performance in 2D and 3D control tasks.
Pose6DAug enables robot data augmentation by swapping objects in successful episodes while preserving physically valid 6D pose trajectories. It operates in 3D using a mesh anchored by temporally coherent poses, ensuring multi-view consistency and physical plausibility. Fine-tuning a VLA policy on this augmented data improves novel object success rates by 16.5% over state-of-the-art baselines.
A new system uses subject-aware prompting to adapt tutoring strategies based on student performance and discipline. A/B testing with 656 student conversations shows the model reduces interactions by 3 turns and increases learning strategy conversion from 19.1% to 28.1% with a stochastic router.
MedRLs enables long-context clinical reasoning by recursively inspecting patient data across text, images, sensors, and guidelines. It integrates specialized agents and a Clinical Evidence Graph Memory to connect observations with evidence and referral criteria, supporting sensor-triggered reasoning and uncertainty-gated clinician review.
MAMO introduces a multi-agent reinforcement learning approach to address the challenge of balancing cost minimization and constraint satisfaction in dynamic environments. It decouples task execution from reward weight selection, treating the choice of weights as a learning problem to enable more autonomous and robust solutions.
TerraMARS is an end-to-end pipeline that uses a domain-adapted small language model to extract structured information from Mars science literature. It converts unstructured text into JSON format and supports Mars terraforming-related question answering, enabling integration into habitability modeling and digital twin applications. The pipeline uses Google Gemma 3 1B fine-tuned with QLoRA on Mars-specific datasets, though further work is needed to improve accuracy and factual consistency.
A study evaluates four AI review systems across six language models, finding OpenAIReview with GPT-5.5 achieves 83.0% accuracy in matching paper quality to external signals and detects 71.6% of injected errors. Real user feedback shows positive sentiment, with a 1.44-to-1 vote ratio, though false positives and minor nitpicks remain common.
AgentFinVQA introduces a multi-agent pipeline for financial chart question answering that ensures auditability and on-premise deployability without significant accuracy loss. It outperforms baseline models by +7.68 pp using a proprietary backbone and +4.84 pp with open-weights Qwen3.6-27B-FP8, while providing a confidence signal via verifier output that improves human review routing.
AtomMem introduces a memory system that stores high-value atomic facts from long-form interactions. It uses hierarchical event structures and temporal profiles to capture coherent episodic contexts and track evolving user attributes, enabling stable and efficient memory evolution. Experiments on the LoCoMo benchmark show AtomMem achieves state-of-the-art performance in reasoning tasks.
A zero-shot agentic workflow using open-source LLMs extracts 13 College of American Pathologists synoptic fields from lung resection pathology reports. The best model (GPT-OSS-20B) achieved a Micro-F1 of 0.893, outperforming baseline recall and accurately capturing complex pathologic relations without task-specific training.
Multi-Agent Transactive Memory (MATM) enables population-level storage and retrieval of agent-generated trajectories. It allows producer agents to share procedural knowledge with consumer agents, improving task performance and reducing interaction steps in interactive environments like ALFWorld and WebArena without coordination or joint training.
GEMS enables training-free superposition of multiple semantic directions in LLMs by addressing distributional deviation and directional interference through geometric constraints. On GSM8K, it maintains 98% accuracy with three non-mathematical directions, while unconstrained addition drops to 4%; on Wikitext-2, it increases PPL by only 2.2%.
A new framework enables large language models to learn 'Connect the Dots' by using reinforcement learning with long rollout sequences. The method includes tailored tasks and environments to foster meta-capability development, showing strong cross-domain generalization and performance in out-of-distribution settings. Implementations are available at https://github.com/agentscope-ai/Trinity-RFT/tree/research/cod/examples/research_cod.
LLM agents commonly select higher-privilege tools despite sufficient lower-privilege alternatives. This over-privileged behavior is amplified by transient tool failures and does not reliably improve with general safety alignment. A new privilege-aware post-training defense reduces unnecessary high-privilege tool use while maintaining agent capabilities.