OThink-MR1: Stimulating multimodal generalized reasoning capabilities through dynamic reinforcement learning

Multimodal Language Models have gained significant traction for their ability to process diverse input data types and generate coherent, contextually relevant outputs across various applications. While supervised fine-tuning (SFT) has been the predominant approach to enhance MLLM capabilities in task-specific optimization, it often falls short in fostering crucial generalized reasoning abilities. Despite the potential of reinforcement learning (RL) to address these limitations, it faces two issues: (1) its generalized capabilities in multimodal tasks remain underexplored. (2) its training constraints such as constant Kullback-Leibler or clamp strategy easily lead to suboptimal bottleneck. To adress these issues, we introduce OThink-MR1, a framework that extends RL to MLLMs, enabling them to achieve deeper understanding and reasoning across multimodal tasks. We design a dynamic Kullback-Leibler strategy that significantly enhances RL performance, surpassing SFT in same-task evaluations. Also, we are the first to reveal that RL exhibits remarkable cross-task generalization capabilities, which shows that models post-trained with RL on one multimodal task can be effectively transfered to another tasks. Finally, extensive experiments demonstrate the great reasoning ability of our proposed OThink-MR1.

PLM: Efficient Peripheral Language Models Hardware-Co-Designed for Ubiquitous Computing

While scaling laws have been continuously validated in large language models (LLMs) with increasing model parameters, the inherent tension between the inference demands of LLMs and the limited resources of edge devices poses a critical challenge to the development of edge intelligence. Recently, numerous small language models have emerged, aiming to distill the capabilities of LLMs into smaller footprints. However, these models often retain the fundamental architectural principles of their larger counterparts, still imposing considerable strain on the storage and bandwidth capacities of edge devices. In this paper, we introduce the PLM, a Peripheral Language Model, developed through a co-design process that jointly optimizes model architecture and edge system constraints. The PLM utilizes a Multi-head Latent Attention mechanism and employs the squared ReLU activation function to encourage sparsity, thereby reducing peak memory footprint during inference. During training, we collect and reorganize open-source datasets, implement a multi-phase training strategy, and empirically investigate the Warmup-Stable-Decay-Constant (WSDC) learning rate scheduler. Additionally, we incorporate Reinforcement Learning from Human Feedback (RLHF) by adopting the ARIES preference learning approach. Following a two-phase SFT process, this method yields performance gains of 2% in general tasks, 9% in the GSM8K task, and 11% in coding tasks. In addition to its novel architecture, evaluation results demonstrate that PLM outperforms existing small language models trained on publicly available data while maintaining the lowest number of activated parameters. Furthermore, deployment across various edge devices, including consumer-grade GPUs, mobile phones, and Raspberry Pis, validates PLM's suitability for peripheral applications. The PLM series models are publicly available at https://github.com/plm-team/PLM.

ST-FlowNet: An Efficient Spiking Neural Network for Event-Based Optical Flow Estimation

Spiking Neural Networks (SNNs) have emerged as a promising tool for event-based optical flow estimation tasks due to their ability to leverage spatio-temporal information and low-power capabilities. However, the performance of SNN models is often constrained, limiting their application in real-world scenarios. In this work, we address this gap by proposing a novel neural network architecture, ST-FlowNet, specifically tailored for optical flow estimation from event-based data. The ST-FlowNet architecture integrates ConvGRU modules to facilitate cross-modal feature augmentation and temporal alignment of the predicted optical flow, improving the network's ability to capture complex motion dynamics. Additionally, to overcome the challenges associated with training SNNs, we introduce a novel approach to derive SNN models from pre-trained artificial neural networks (ANNs) through ANN-to-SNN conversion or our proposed BISNN method. Notably, the BISNN method alleviates the complexities involved in biological parameter selection, further enhancing the robustness of SNNs in optical flow estimation tasks. Extensive evaluations on three benchmark event-based datasets demonstrate that the SNN-based ST-FlowNet model outperforms state-of-the-art methods, delivering superior performance in accurate optical flow estimation across a diverse range of dynamic visual scenes. Furthermore, the inherent energy efficiency of SNN models is highlighted, establishing a compelling advantage for their practical deployment. Overall, our work presents a novel framework for optical flow estimation using SNNs and event-based data, contributing to the advancement of neuromorphic vision applications.

ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning

Recent research on Reasoning of Large Language Models (LLMs) has sought to further enhance their performance by integrating meta-thinking -- enabling models to monitor, evaluate, and control their reasoning processes for more adaptive and effective problem-solving. However, current single-agent work lacks a specialized design for acquiring meta-thinking, resulting in low efficacy. To address this challenge, we introduce Reinforced Meta-thinking Agents (ReMA), a novel framework that leverages Multi-Agent Reinforcement Learning (MARL) to elicit meta-thinking behaviors, encouraging LLMs to think about thinking. ReMA decouples the reasoning process into two hierarchical agents: a high-level meta-thinking agent responsible for generating strategic oversight and plans, and a low-level reasoning agent for detailed executions. Through iterative reinforcement learning with aligned objectives, these agents explore and learn collaboration, leading to improved generalization and robustness. Experimental results demonstrate that ReMA outperforms single-agent RL baselines on complex reasoning tasks, including competitive-level mathematical benchmarks and LLM-as-a-Judge benchmarks. Comprehensive ablation studies further illustrate the evolving dynamics of each distinct agent, providing valuable insights into how the meta-thinking reasoning process enhances the reasoning capabilities of LLMs.

Bioinspired Sensing of Undulatory Flow Fields Generated by Leg Kicks in Swimming

The artificial lateral line (ALL) is a bioinspired flow sensing system for underwater robots, comprising of distributed flow sensors. The ALL has been successfully applied to detect the undulatory flow fields generated by body undulation and tail-flapping of bioinspired robotic fish. However, its feasibility and performance in sensing the undulatory flow fields produced by human leg kicks during swimming has not been systematically tested and studied. This paper presents a novel sensing framework to investigate the undulatory flow field generated by swimmer's leg kicks, leveraging bioinspired ALL sensing. To evaluate the feasibility of using the ALL system for sensing the undulatory flow fields generated by swimmer leg kicks, this paper designs an experimental platform integrating an ALL system and a lab-fabricated human leg model. To enhance the accuracy of flow sensing, this paper proposes a feature extraction method that dynamically fuses time-domain and time-frequency characteristics. Specifically, time-domain features are extracted using one-dimensional convolutional neural networks and bidirectional long short-term memory networks (1DCNN-BiLSTM), while time-frequency features are extracted using short-term Fourier transform and two-dimensional convolutional neural networks (STFT-2DCNN). These features are then dynamically fused based on attention mechanisms to achieve accurate sensing of the undulatory flow field. Furthermore, extensive experiments are conducted to test various scenarios inspired by human swimming, such as leg kick pattern recognition and kicking leg localization, achieving satisfactory results.

Embodied multi-modal sensing with a soft modular arm powered by physical reservoir computing

Soft robots have become increasingly popular for complex manipulation tasks requiring gentle and safe contact. However, their softness makes accurate control challenging, and high-fidelity sensing is a prerequisite to adequate control performance. To this end, many flexible and embedded sensors have been created over the past decade, but they inevitably increase the robot's complexity and stiffness. This study demonstrates a novel approach that uses simple bending strain gauges embedded inside a modular arm to extract complex information regarding its deformation and working conditions. The core idea is based on physical reservoir computing (PRC): A soft body's rich nonlinear dynamic responses, captured by the inter-connected bending sensor network, could be utilized for complex multi-modal sensing with a simple linear regression algorithm. Our results show that the soft modular arm reservoir can accurately predict body posture (bending angle), estimate payload weight, determine payload orientation, and even differentiate two payloads with only minimal difference in weight -- all using minimal digital computing power.

Breaking Free from MMI: A New Frontier in Rationalization by Probing Input Utilization

Extracting a small subset of crucial rationales from the full input is a key problem in explainability research. The most widely used fundamental criterion for rationale extraction is the maximum mutual information (MMI) criterion. In this paper, we first demonstrate that MMI suffers from diminishing marginal returns. Once part of the rationale has been identified, finding the remaining portions contributes only marginally to increasing the mutual information, making it difficult to use MMI to locate the rest. In contrast to MMI that aims to reproduce the prediction, we seek to identify the parts of the input that the network can actually utilize. This is achieved by comparing how different rationale candidates match the capability space of the weight matrix. The weight matrix of a neural network is typically low-rank, meaning that the linear combinations of its column vectors can only cover part of the directions in a high-dimensional space (high-dimension: the dimensions of an input vector). If an input is fully utilized by the network, {it generally matches these directions (e.g., a portion of a hypersphere), resulting in a representation with a high norm. Conversely, if an input primarily falls outside (orthogonal to) these directions}, its representation norm will approach zero, behaving like noise that the network cannot effectively utilize. Building on this, we propose using the norms of rationale candidates as an alternative objective to MMI. Through experiments on four text classification datasets and one graph classification dataset using three network architectures (GRUs, BERT, and GCN), we show that our method outperforms MMI and its improved variants in identifying better rationales. We also compare our method with a representative LLM (llama-3.1-8b-instruct) and find that our simple method gets comparable results to it and can sometimes even outperform it.

DB-Explore: Automated Database Exploration and Instruction Synthesis for Text-to-SQL

Recent text-to-SQL systems powered by large language models (LLMs) have demonstrated remarkable performance in translating natural language queries into SQL. However, these systems often struggle with complex database structures and domain-specific queries, as they primarily focus on enhancing logical reasoning and SQL syntax while overlooking the critical need for comprehensive database understanding. To address this limitation, we propose DB-Explore, a novel framework that systematically aligns LLMs with database knowledge through automated exploration and instruction synthesis. DB-Explore constructs database graphs to capture complex relational schemas, leverages GPT-4 to systematically mine structural patterns and semantic knowledge, and synthesizes instructions to distill this knowledge for efficient fine-tuning of LLMs. Our framework enables comprehensive database understanding through diverse sampling strategies and automated instruction generation, bridging the gap between database structures and language models. Experiments conducted on the SPIDER and BIRD benchmarks validate the effectiveness of DB-Explore, achieving an execution accuracy of 52.1% on BIRD and 84.0% on SPIDER. Notably, our open-source implementation, based on the Qwen2.5-coder-7B model, outperforms multiple GPT-4-driven text-to-SQL systems in comparative evaluations, and achieves near state-of-the-art performance with minimal computational cost.

AskToAct: Enhancing LLMs Tool Use via Self-Correcting Clarification

Large language models (LLMs) have demonstrated remarkable capabilities in tool learning. In real-world scenarios, user queries are often ambiguous and incomplete, requiring effective clarification. However, existing interactive clarification approaches face two critical limitations: reliance on manually constructed datasets and lack of error correction mechanisms during multi-turn clarification. We present AskToAct, which addresses these challenges by exploiting the structural mapping between queries and their tool invocation solutions. Our key insight is that tool parameters naturally represent explicit user intents. By systematically removing key parameters from queries while retaining them as ground truth, we enable automated construction of high-quality training data. We further enhance model robustness by fine-tuning on error-correction augmented data using selective masking mechanism, enabling dynamic error detection during clarification interactions. Comprehensive experiments demonstrate that AskToAct significantly outperforms existing approaches, achieving above 79% accuracy in recovering critical unspecified intents and enhancing clarification efficiency by an average of 48.34% while maintaining high accuracy in tool invocation. Our framework exhibits robust performance across varying complexity levels and successfully generalizes to entirely unseen APIs without additional training, achieving performance comparable to GPT-4 with substantially fewer computational resources.

Noise-Injected Spiking Graph Convolution for Energy-Efficient 3D Point Cloud Denoising

Spiking neural networks (SNNs), inspired by the spiking computation paradigm of the biological neural systems, have exhibited superior energy efficiency in 2D classification tasks over traditional artificial neural networks (ANNs). However, the regression potential of SNNs has not been well explored, especially in 3D point cloud processing.In this paper, we propose noise-injected spiking graph convolutional networks to leverage the full regression potential of SNNs in 3D point cloud denoising. Specifically, we first emulate the noise-injected neuronal dynamics to build noise-injected spiking neurons. On this basis, we design noise-injected spiking graph convolution for promoting disturbance-aware spiking representation learning on 3D points. Starting from the spiking graph convolution, we build two SNN-based denoising networks. One is a purely spiking graph convolutional network, which achieves low accuracy loss compared with some ANN-based alternatives, while resulting in significantly reduced energy consumption on two benchmark datasets, PU-Net and PC-Net. The other is a hybrid architecture that combines ANN-based learning with a high performance-efficiency trade-off in just a few time steps. Our work lights up SNN's potential for 3D point cloud denoising, injecting new perspectives of exploring the deployment on neuromorphic chips while paving the way for developing energy-efficient 3D data acquisition devices.