Human Feedback and Eval Paper Explorer

Match reason: Matches selected tags (Demonstrations).

• On real-world animal decision-making tasks, ICL achieves competitive performance with models designed by human experts.

Match reason: Matches selected tags (Demonstrations).

• Experiments across diverse benchmarks and model architectures demonstrate that weak reasoners can effectively incentivize reasoning in stronger student models, consistently improving performance across a wide range of reasoning tasks.

Match reason: Matches selected tags (Demonstrations).

• We introduce PC Agent-E, an efficient agent training framework that significantly reduces reliance on large-scale human demonstrations.
• Trained on these enriched trajectories, our PC Agent-E model achieved a remarkable 141 relative improvement, and even surpassed the Claude 3.7 Sonnet by 10% in relative terms on WindowsAgentArena-V2, an improved benchmark we also released.

Match reason: Matches selected tags (Demonstrations).

• Large language models (LLMs) exhibit strong capabilities as decision-making agents by interleaving reasoning and actions, as seen in ReAct-style frameworks.
• We propose Structured Agent Distillation, a framework that compresses large LLM-based agents into smaller student models while preserving both reasoning fidelity and action consistency.

Match reason: Matches selected tags (Demonstrations).

Match reason: Matches selected tags (Demonstrations).

• The task is turn-based, multi-agent, and physically grounded, posing significant challenges due to its long-horizon dependencies, tight inter-agent coupling, and the underactuated dynamics of quadrotors.

Match reason: Matches selected tags (Demonstrations).

Match reason: Matches selected tags (Demonstrations).

• In this work we present a framework for robots to continually learn tasks and visuo-motor skills and query for novel skills via dialog interactions with human users.
• Moreover, with our IRB approved human-subjects study we demonstrate that our dialog based continual learning framework allows users to teach robots cooking skills successfully (100%) while spending a higher ratio of time on finishing an…

Match reason: Matches selected tags (Demonstrations).

• We provide a comprehensive suite of tasks ranging from single-drone drills to multi-drone cooperative and competitive tasks, accompanied by baseline evaluations of representative reinforcement learning (RL), multi-agent reinforcement…
• Simulation results show that on-policy RL methods outperform off-policy methods in single-agent tasks, but both approaches struggle in complex tasks that combine motion control and strategic play.

Match reason: Matches selected tags (Demonstrations).

• We propose CowPilot, a framework supporting autonomous as well as human-agent collaborative web navigation, and evaluation across task success and task efficiency.
• We conducted case studies on five common websites and found that the human-agent collaborative mode achieves the highest success rate of 95% while requiring humans to perform only 15.2% of the total steps.

Match reason: Matches selected tags (Demonstrations).

• We evaluate our method on 12 publicly available benchmarks spanning instruction-following, mathematical reasoning, and commonsense reasoning, across six base models.
• Our results show that models trained with pseudo-code follow instructions more reliably, achieving relative gains of 8-21\% on instruction following benchmarks, while largely preserving and in some cases improving performance on…

Match reason: Matches selected tags (Demonstrations).

• Inspired by human metacognitive learning and creative problem-solving, we address this limitation by exploring a fundamental question: Can LLMs be empowered with metacognitive capabilities to reason, reflect, and create, thereby enhancing…
• We propose a more challenging robotic benchmark task and evaluate our framework on the existing benchmark and the novel task.

Match reason: Matches selected tags (Demonstrations).

• Existing techniques are either reward-based (training a reward model on preference pairs and optimizing with reinforcement learning) or reward-free (directly fine-tuning on ranked outputs).
• Recent research shows that well-tuned reward-based pipelines remain the most robust, and single-response demonstrations can outperform pairwise preference data.

Match reason: Matches selected tags (Demonstrations).

Match reason: Matches selected tags (Demonstrations).