Unlocking Efficient Long-to-Short LLM Reasoning with Model Merging

The transition from System 1 to System 2 reasoning in large language models (LLMs) has marked significant advancements in handling complex tasks through deliberate, iterative thinking. However, this progress often comes at the cost of efficiency, as models tend to overthink, generating redundant reasoning steps without proportional improvements in output quality. Long-to-Short (L2S) reasoning has emerged as a promising solution to this challenge, aiming to balance reasoning depth with practical efficiency. While existing approaches, such as supervised fine-tuning (SFT), reinforcement learning (RL), and prompt engineering, have shown potential, they are either computationally expensive or unstable. Model merging, on the other hand, offers a cost-effective and robust alternative by integrating the quick-thinking capabilities of System 1 models with the methodical reasoning of System 2 models. In this work, we present a comprehensive empirical study on model merging for L2S reasoning, exploring diverse methodologies, including task-vector-based, SVD-based, and activation-informed merging. Our experiments reveal that model merging can reduce average response length by up to 55% while preserving or even improving baseline performance. We also identify a strong correlation between model scale and merging efficacy with extensive evaluations on 1.5B/7B/14B/32B models. Furthermore, we investigate the merged model's ability to self-critique and self-correct, as well as its adaptive response length based on task complexity. Our findings highlight model merging as a highly efficient and effective paradigm for L2S reasoning, offering a practical solution to the overthinking problem while maintaining the robustness of System 2 reasoning. This work can be found on Github https://github.com/hahahawu/Long-to-Short-via-Model-Merging.

Pig behavior dataset and Spatial-temporal perception and enhancement networks based on the attention mechanism for pig behavior recognition

The recognition of pig behavior plays a crucial role in smart farming and welfare assurance for pigs. Currently, in the field of pig behavior recognition, the lack of publicly available behavioral datasets not only limits the development of innovative algorithms but also hampers model robustness and algorithm optimization.This paper proposes a dataset containing 13 pig behaviors that significantly impact welfare.Based on this dataset, this paper proposes a spatial-temporal perception and enhancement networks based on the attention mechanism to model the spatiotemporal features of pig behaviors and their associated interaction areas in video data. The network is composed of a spatiotemporal perception network and a spatiotemporal feature enhancement network. The spatiotemporal perception network is responsible for establishing connections between the pigs and the key regions of their behaviors in the video data. The spatiotemporal feature enhancement network further strengthens the important spatial features of individual pigs and captures the long-term dependencies of the spatiotemporal features of individual behaviors by remodeling these connections, thereby enhancing the model's perception of spatiotemporal changes in pig behaviors. Experimental results demonstrate that on the dataset established in this paper, our proposed model achieves a MAP score of 75.92%, which is an 8.17% improvement over the best-performing traditional model. This study not only improces the accuracy and generalizability of individual pig behavior recognition but also provides new technological tools for modern smart farming. The dataset and related code will be made publicly available alongside this paper.

A Helping (Human) Hand in Kinematic Structure Estimation

Visual uncertainties such as occlusions, lack of texture, and noise present significant challenges in obtaining accurate kinematic models for safe robotic manipulation. We introduce a probabilistic real-time approach that leverages the human hand as a prior to mitigate these uncertainties. By tracking the constrained motion of the human hand during manipulation and explicitly modeling uncertainties in visual observations, our method reliably estimates an object's kinematic model online. We validate our approach on a novel dataset featuring challenging objects that are occluded during manipulation and offer limited articulations for perception. The results demonstrate that by incorporating an appropriate prior and explicitly accounting for uncertainties, our method produces accurate estimates, outperforming two recent baselines by 195% and 140%, respectively. Furthermore, we demonstrate that our approach's estimates are precise enough to allow a robot to manipulate even small objects safely.

AttentionPredictor: Temporal Pattern Matters for Efficient LLM Inference

With the development of large language models (LLMs), efficient inference through Key-Value (KV) cache compression has attracted considerable attention, especially for long-context generation. To compress the KV cache, recent methods identify critical KV tokens through heuristic ranking with attention scores. However, these methods often struggle to accurately determine critical tokens as they neglect the \textit{temporal patterns} in attention scores, resulting in a noticeable degradation in LLM performance. To address this challenge, we propose AttentionPredictor, which is the first learning-based critical token identification approach. Specifically, AttentionPredictor learns a lightweight convolution model to capture spatiotemporal patterns and predict the next-token attention score. An appealing feature of AttentionPredictor is that it accurately predicts the attention score while consuming negligible memory. Moreover, we propose a cross-token critical cache prefetching framework that hides the token estimation time overhead to accelerate the decoding stage. By retaining most of the attention information, AttentionPredictor achieves 16$\times$ KV cache compression with comparable LLM performance, significantly outperforming the state-of-the-art.

KVTuner: Sensitivity-Aware Layer-wise Mixed Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference

KV cache quantization can improve Large Language Models (LLMs) inference throughput and latency in long contexts and large batch-size scenarios while preserving LLMs effectiveness. However, current methods have three unsolved issues: overlooking layer-wise sensitivity to KV cache quantization, high overhead of online fine-grained decision-making, and low flexibility to different LLMs and constraints. Therefore, we thoroughly analyze the inherent correlation of layer-wise transformer attention patterns to KV cache quantization errors and study why key cache is more important than value cache for quantization error reduction. We further propose a simple yet effective framework KVTuner to adaptively search for the optimal hardware-friendly layer-wise KV quantization precision pairs for coarse-grained KV cache with multi-objective optimization and directly utilize the offline searched configurations during online inference. To reduce the computational cost of offline calibration, we utilize the intra-layer KV precision pair pruning and inter-layer clustering to reduce the search space. Experimental results show that we can achieve nearly lossless 3.25-bit mixed precision KV cache quantization for LLMs like Llama-3.1-8B-Instruct and 4.0-bit for sensitive models like Qwen2.5-7B-Instruct on mathematical reasoning tasks. The maximum inference throughput can be improved by 38.3% compared with KV8 quantization over various context lengths.

SeaDAG: Semi-autoregressive Diffusion for Conditional Directed Acyclic Graph Generation

We introduce SeaDAG, a semi-autoregressive diffusion model for conditional generation of Directed Acyclic Graphs (DAGs). Considering their inherent layer-wise structure, we simulate layer-wise autoregressive generation by designing different denoising speed for different layers. Unlike conventional autoregressive generation that lacks a global graph structure view, our method maintains a complete graph structure at each diffusion step, enabling operations such as property control that require the full graph structure. Leveraging this capability, we evaluate the DAG properties during training by employing a graph property decoder. We explicitly train the model to learn graph conditioning with a condition loss, which enhances the diffusion model's capacity to generate graphs that are both realistic and aligned with specified properties. We evaluate our method on two representative conditional DAG generation tasks: (1) circuit generation from truth tables, where precise DAG structures are crucial for realizing circuit functionality, and (2) molecule generation based on quantum properties. Our approach demonstrates promising results, generating high-quality and realistic DAGs that closely align with given conditions.

Logic Synthesis Optimization with Predictive Self-Supervision via Causal Transformers

Contemporary hardware design benefits from the abstraction provided by high-level logic gates, streamlining the implementation of logic circuits. Logic Synthesis Optimization (LSO) operates at one level of abstraction within the Electronic Design Automation (EDA) workflow, targeting improvements in logic circuits with respect to performance metrics such as size and speed in the final layout. Recent trends in the field show a growing interest in leveraging Machine Learning (ML) for EDA, notably through ML-guided logic synthesis utilizing policy-based Reinforcement Learning (RL) methods.Despite these advancements, existing models face challenges such as overfitting and limited generalization, attributed to constrained public circuits and the expressiveness limitations of graph encoders. To address these hurdles, and tackle data scarcity issues, we introduce LSOformer, a novel approach harnessing Autoregressive transformer models and predictive SSL to predict the trajectory of Quality of Results (QoR). LSOformer integrates cross-attention modules to merge insights from circuit graphs and optimization sequences, thereby enhancing prediction accuracy for QoR metrics. Experimental studies validate the effectiveness of LSOformer, showcasing its superior performance over baseline architectures in QoR prediction tasks, where it achieves improvements of 5.74%, 4.35%, and 17.06% on the EPFL, OABCD, and proprietary circuits datasets, respectively, in inductive setup.

KAN-HyperpointNet for Point Cloud Sequence-Based 3D Human Action Recognition

Point cloud sequence-based 3D action recognition has achieved impressive performance and efficiency. However, existing point cloud sequence modeling methods cannot adequately balance the precision of limb micro-movements with the integrity of posture macro-structure, leading to the loss of crucial information cues in action inference. To overcome this limitation, we introduce D-Hyperpoint, a novel data type generated through a D-Hyperpoint Embedding module. D-Hyperpoint encapsulates both regional-momentary motion and global-static posture, effectively summarizing the unit human action at each moment. In addition, we present a D-Hyperpoint KANsMixer module, which is recursively applied to nested groupings of D-Hyperpoints to learn the action discrimination information and creatively integrates Kolmogorov-Arnold Networks (KAN) to enhance spatio-temporal interaction within D-Hyperpoints. Finally, we propose KAN-HyperpointNet, a spatio-temporal decoupled network architecture for 3D action recognition. Extensive experiments on two public datasets: MSR Action3D and NTU-RGB+D 60, demonstrate the state-of-the-art performance of our method.

MTLSO: A Multi-Task Learning Approach for Logic Synthesis Optimization

Electronic Design Automation (EDA) is essential for IC design and has recently benefited from AI-based techniques to improve efficiency. Logic synthesis, a key EDA stage, transforms high-level hardware descriptions into optimized netlists. Recent research has employed machine learning to predict Quality of Results (QoR) for pairs of And-Inverter Graphs (AIGs) and synthesis recipes. However, the severe scarcity of data due to a very limited number of available AIGs results in overfitting, significantly hindering performance. Additionally, the complexity and large number of nodes in AIGs make plain GNNs less effective for learning expressive graph-level representations. To tackle these challenges, we propose MTLSO - a Multi-Task Learning approach for Logic Synthesis Optimization. On one hand, it maximizes the use of limited data by training the model across different tasks. This includes introducing an auxiliary task of binary multi-label graph classification alongside the primary regression task, allowing the model to benefit from diverse supervision sources. On the other hand, we employ a hierarchical graph representation learning strategy to improve the model's capacity for learning expressive graph-level representations of large AIGs, surpassing traditional plain GNNs. Extensive experiments across multiple datasets and against state-of-the-art baselines demonstrate the superiority of our method, achieving an average performance gain of 8.22\% for delay and 5.95\% for area.

A Biologically Inspired Design Principle for Building Robust Robotic Systems

Robustness, the ability of a system to maintain performance under significant and unanticipated environmental changes, is a critical property for robotic systems. While biological systems naturally exhibit robustness, there is no comprehensive understanding of how to achieve similar robustness in robotic systems. In this work, we draw inspirations from biological systems and propose a design principle that advocates active interconnections among system components to enhance robustness to environmental variations. We evaluate this design principle in a challenging long-horizon manipulation task: solving lockboxes. Our extensive simulated and real-world experiments demonstrate that we could enhance robustness against environmental changes by establishing active interconnections among system components without substantial changes in individual components. Our findings suggest that a systematic investigation of design principles in system building is necessary. It also advocates for interdisciplinary collaborations to explore and evaluate additional principles of biological robustness to advance the development of intelligent and adaptable robotic systems.