AI Agent Papers2026
Provides a terminal-style benchmark of 46 long-horizon tasks decomposed into fine-grained graded subtasks to produce dense intermediate rewards and partial credit, enabling evaluation of long-horizon planning, long-context management, and iterative debugging. Tasks typically require hundreds of episodes and minutes-to-hours of execution; baseline evaluations report high token and episode consumption with low pass rates, highlighting evaluation headroom.
AI Dataset2026
Simulates a hospital LIMS to benchmark agentic clinical reasoning: agents inspect demographics, medications, lab orders/results and then submit ICD‑10 diagnostic reports scored by deterministic, context‑aware graders. Ships as an OpenEnv/FastAPI runtime with 8 scenarios, step‑level rewards and trajectory capture for RL, tool‑use and evaluation.
AI Dataset2026
Evaluates agents inside a structured hospital workflow via a downloadable FastAPI runtime that enforces role-specific tool permissions, evidence-before-treatment discipline, deterministic grading, dense process rewards, and full trajectory logging. Designed for RL, offline policy learning, multi-agent workflow research and process-supervision datasets; not for real patient care.
Proposes Riemannian Isometric Policy Optimization (RIPO) to fix exploration collapse in PPO-style RL for LLMs by aligning policy updates with the policy manifold's Riemannian geometry, improving exploration–exploitation balance and optimization stability across competition benchmarks.
AI Agent Papers2026
Unifies high-level visual-language reasoning and low-level control for visual navigation by decoupling cognition and control: a slow vision-language reasoner produces pixel goals with explicit chain-of-thought, and a fast action expert converts those anchors into continuous waypoints for robust urban and indoor navigation.
Uses pretrained multimodal LLMs as zero-shot, training-free reward models for text-to-image RL by scoring how well the original text prompt can be recovered from a generated image via image-conditioned prompt log-likelihood; includes a Self-SpectraReward closed-loop variant.
AI Agent Papers2026
Converts completed on-policy trajectories into natural-language 'hindsight skills' and converts the skill-induced action probability shifts into a dense token-level on-policy distillation signal, jointly optimized with outcome-based RL to improve sample efficiency and long-horizon agent behavior.
Embodied AI2026
Analyzes adversarial weaknesses of World-Action Models (WAMs) via BadWAM, a framework that crafts visual perturbations to decouple a model’s imagined future from its executed actions. Introduces two attack modes—action-only (disruptive) and imagination-preserving (stealthy)—and shows large drops in closed-loop task success (e.g., 96.5%→43.1%).
Enables RL post-training with million-token prompts under a fixed GPU budget by evaluating shared prompt state without autograd, retaining only minimal model state, and replaying short response branches; instantiated as GRPO and demonstrated on Qwen3.6-27B and GLM-5.2 up to multi-million token execution.
AI Model2026
1B-parameter text-generation model tuned for conversational and agentic workflows with a focus on security and vulnerability-detection; suited for low-cost or on-prem/edge deployments and terminal-agent integrations.
A PyTorch-native training framework for agentic reinforcement learning research that keeps researcher-facing code compact and editable. Uses an asynchronous loop to train multimodal and mixture-of-experts policies while never training on tokens the agent didn't generate; matches Megatron-style stacks under a comparable protocol and ships recipes and containers on GitHub.
Provides a unified survey of progress-reward modeling for robotic learning, detailing interfaces, modeling techniques, and evaluation practices. Organizes the literature into three perspectives—interface, model internals, and data/benchmarks—and highlights limitations and open problems. Useful for researchers designing rewards for long-horizon or sparse-reward robotic tasks.