Think Thrice Before You Act: Progressive Thought Refinement in Large Language Models

Recent advancements in large language models (LLMs) have demonstrated that progressive refinement, rather than providing a single answer, results in more accurate and thoughtful outputs. However, existing methods often rely heavily on supervision signals to evaluate previous responses, making it difficult to assess output quality in more open-ended scenarios effectively. Additionally, these methods are typically designed for specific tasks, which limits their generalization to new domains. To address these limitations, we propose Progressive Thought Refinement (PTR), a framework that enables LLMs to refine their responses progressively. PTR operates in two phases: (1) Thought data construction stage: We propose a weak and strong model collaborative selection strategy to build a high-quality progressive refinement dataset to ensure logical consistency from thought to answers, and the answers are gradually refined in each round. (2) Thought-Mask Fine-Tuning Phase: We design a training structure to mask the "thought" and adjust loss weights to encourage LLMs to refine prior thought, teaching them to implicitly understand "how to improve" rather than "what is correct." Experimental results show that PTR significantly enhances LLM performance across ten diverse tasks (avg. from 49.6% to 53.5%) without task-specific fine-tuning. Notably, in more open-ended tasks, LLMs also demonstrate substantial improvements in the quality of responses beyond mere accuracy, suggesting that PTR truly teaches LLMs to self-improve over time.

Impact of Tactile Sensor Quantities and Placements on Learning-based Dexterous Manipulation

Tactile information effectively enables faster training and better task performance for learning-based in-hand manipulation. Existing approaches are validated in simulated environments with a large number of tactile sensors. However, attaching such sensors to a real robot hand is not applicable due to high cost and physical limitations. To enable real-world adoption of tactile sensors, this study investigates the impact of tactile sensors, including their varying quantities and placements on robot hands, on the dexterous manipulation task performance and analyzes the importance of each. Through empirically decreasing the sensor quantities, we successfully find an optimized set of tactile sensors (21 sensors) configuration, which keeps over 93% task performance with only 20% sensor quantities compared to the original set (92 sensors) for the block manipulation task, leading to a potential reduction of over 80% in sensor manufacturing and design costs. To transform the empirical results into a generalizable understanding, we build a task performance prediction model with a weighted linear regression algorithm and use it to forecast the task performance with different sensor configurations. To show its generalizability, we verified this model in egg and pen manipulation tasks and achieved an average prediction error of 3.12%.

Bi-directional Momentum-based Haptic Feedback and Control System for Dexterous Telemanipulation

Haptic feedback is essential for dexterous telemanipulation that enables operators to control robotic hands remotely with high skill and precision, mimicking a human hand's natural movement and sensation. However, current haptic methods for dexterous telemanipulation cannot support torque feedback, resulting in object rotation and rolling mismatches. The operator must make tedious adjustments in these tasks, leading to delays, reduced situational awareness, and suboptimal task performance. This work presents a Bi-directional Momentum-based Haptic Feedback and Control (Bi-Hap) system for real-time dexterous telemanipulation. Bi-Hap integrates multi-modal sensors to extract human interactive information with the object and share it with the robot's learning-based controller. A Field-Oriented Control (FOC) algorithm is developed to enable the integrated brushless active momentum wheel to generate precise torque and vibrative feedback, bridging the gap between human intent and robotic actions. Different feedback strategies are designed for varying error states to align with the operator's intuition. Extensive experiments with human subjects using a virtual Shadow Dexterous Hand demonstrate the effectiveness of Bi-Hap in enhancing task performance and user confidence. Bi-Hap achieved real-time feedback capability with low command following latency (delay<0.025s) and highly accurate torque feedback (RMSE<0.010 Nm).

Agent Group Chat: An Interactive Group Chat Simulacra For Better Eliciting Collective Emergent Behavior

To investigate the role of language in human collective behaviors, we developed the Agent Group Chat simulation to simulate linguistic interactions among multi-agent in different settings. Agents are asked to free chat in this simulation for their own purposes based on their character setting, aiming to see agents exhibit emergent behaviours that are both unforeseen and significant. Four narrative scenarios, Inheritance Disputes, Law Court Debates, Philosophical Discourses, Movie Casting Contention, are integrated into Agent Group Chat to evaluate its support for diverse storylines. By configuring specific environmental settings within Agent Group Chat, we are able to assess whether agents exhibit behaviors that align with human expectations. We evaluate the disorder within the environment by computing the n-gram Shannon entropy of all the content speak by characters. Our findings reveal that under the premise of agents possessing substantial alignment with human expectations, facilitating more extensive information exchange within the simulation ensures greater orderliness amidst diversity, which leads to the emergence of more unexpected and meaningful emergent behaviors. The code is open source in https://github.com/MikeGu721/AgentGroup, and online platform will be open soon.

CO-SNE: Dimensionality Reduction and Visualization for Hyperbolic Data

Hyperbolic space can embed tree metric with little distortion, a desirable property for modeling hierarchical structures of real-world data and semantics. While high-dimensional embeddings often lead to better representations, most hyperbolic models utilize low-dimensional embeddings, due to non-trivial optimization as well as the lack of a visualization for high-dimensional hyperbolic data. We propose CO-SNE, extending the Euclidean space visualization tool, t-SNE, to hyperbolic space. Like t-SNE, it converts distances between data points to joint probabilities and tries to minimize the Kullback-Leibler divergence between the joint probabilities of high-dimensional data $X$ and low-dimensional embeddings $Y$. However, unlike Euclidean space, hyperbolic space is inhomogeneous: a volume could contain a lot more points at a location far from the origin. CO-SNE thus uses hyperbolic normal distributions for $X$ and hyberbolic \underline{C}auchy instead of t-SNE's Student's t-distribution for $Y$, and it additionally attempts to preserve $X$'s individual distances to the \underline{O}rigin in $Y$. We apply CO-SNE to high-dimensional hyperbolic biological data as well as unsupervisedly learned hyperbolic representations. Our results demonstrate that CO-SNE deflates high-dimensional hyperbolic data into a low-dimensional space without losing their hyperbolic characteristics, significantly outperforming popular visualization tools such as PCA, t-SNE, UMAP, and HoroPCA, the last of which is specifically designed for hyperbolic data.