arXiv subjects

Chenyang Zhao

Publications and source records attributed to Chenyang Zhao.

At least 55 records · Page 3Linked to original sources

arXiv2025

Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space

Human cognition typically involves thinking through abstract, fluid concepts rather than strictly using discrete linguistic tokens. Current reasoning models, however, are constrained to reasoning within the boundaries of human language, processing discrete token embeddings that represent fixed points in the semantic space. This discrete constraint restricts the expressive power and upper potential of such reasoning models, often causing incomplete exploration of reasoning paths, as standard Chain-of-Thought (CoT) methods rely on sampling one token per step. In this work, we introduce Soft Thinking, a training-free method that emulates human-like "soft" reasoning by generating soft, abstract concept tokens in a continuous concept space. These concept tokens are created by the probability-weighted mixture of token embeddings, which form the continuous concept space, enabling smooth transitions and richer representations that transcend traditional discrete boundaries. In essence, each generated concept token encapsulates multiple meanings from related discrete tokens, implicitly exploring various reasoning paths to converge effectively toward the correct answer. Empirical evaluations on diverse mathematical and coding benchmarks consistently demonstrate the effectiveness and efficiency of Soft Thinking, improving pass@1 accuracy by up to 2.48 points while simultaneously reducing token usage by up to 22.4% compared to standard CoT. Qualitative analysis further reveals that Soft Thinking outputs remain highly interpretable and readable, highlighting the potential of Soft Thinking to break the inherent bottleneck of discrete language-based reasoning. Code is available at https://github.com/eric-ai-lab/Soft-Thinking.

cs.CL↗

arXiv2025

Density-based Object Detection in Crowded Scenes

Compared with the generic scenes, crowded scenes contain highly-overlapped instances, which result in: 1) more ambiguous anchors during training of object detectors, and 2) more predictions are likely to be mistakenly suppressed in post-processing during inference. To address these problems, we propose two new strategies, density-guided anchors (DGA) and density-guided NMS (DG-NMS), which uses object density maps to jointly compute optimal anchor assignments and reweighing, as well as an adaptive NMS. Concretely, based on an unbalanced optimal transport (UOT) problem, the density owned by each ground-truth object is transported to each anchor position at a minimal transport cost. And density on anchors comprises an instance-specific density distribution, from which DGA decodes the optimal anchor assignment and re-weighting strategy. Meanwhile, DG-NMS utilizes the predicted density map to adaptively adjust the NMS threshold to reduce mistaken suppressions. In the UOT, a novel overlap-aware transport cost is specifically designed for ambiguous anchors caused by overlapped neighboring objects. Extensive experiments on the challenging CrowdHuman dataset with Citypersons dataset demonstrate that our proposed density-guided detector is effective and robust to crowdedness. The code and pre-trained models will be made available later.

cs.CV↗

arXiv2025

Revisiting Training-Inference Trigger Intensity in Backdoor Attacks

Backdoor attacks typically place a specific trigger on certain training data, such that the model makes prediction errors on inputs with that trigger during inference. Despite the core role of the trigger, existing studies have commonly believed a perfect match between training-inference triggers is optimal. In this paper, for the first time, we systematically explore the training-inference trigger relation, particularly focusing on their mismatch, based on a Training-Inference Trigger Intensity Manipulation (TITIM) workflow. TITIM specifically investigates the training-inference trigger intensity, such as the size or the opacity of a trigger, and reveals new insights into trigger generalization and overfitting. These new insights challenge the above common belief by demonstrating that the training-inference trigger mismatch can facilitate attacks in two practical scenarios, posing more significant security threats than previously thought. First, when the inference trigger is fixed, using training triggers with mixed intensities leads to stronger attacks than using any single intensity. For example, on CIFAR-10 with ResNet-18, mixing training triggers with 1.0 and 0.1 opacities improves the worst-case attack success rate (ASR) (over different testing opacities) of the best single-opacity attack from 10.61\% to 92.77\%. Second, intentionally using certain mismatched training-inference triggers can improve the attack stealthiness, i.e., better bypassing defenses. For example, compared to the training/inference intensity of 1.0/1.0, using 1.0/0.7 decreases the area under the curve (AUC) of the Scale-Up defense from 0.96 to 0.62, while maintaining a high attack ASR (99.65\% vs. 91.62\%). The above new insights are validated to be generalizable across different backdoor attacks, models, datasets, tasks, and (digital/physical) domains.

cs.CR↗

arXiv2025

Evolution of cooperation in a bimodal mixture of conditional cooperators

Extensive behavioral experiments reveal that conditional cooperation is a prevalent phenomenon. Previous game-theoretical studies have predominantly relied on hard-manner models, where cooperation is triggered only upon reaching a specific threshold. However, this approach contrasts with the observed flexibility of human behaviors, where individuals adapt their strategies dynamically based on their surroundings. To capture this adaptability, we introduce a soft form of conditional cooperation by integrating the Q-learning algorithm from reinforcement learning. In this form, players not only reciprocate mutual cooperation but may also defect in highly cooperative environments or cooperate in less cooperative settings to maximize rewards. To explore the effects of hard and soft conditional cooperators, we examine their interactions in two scenarios: structural mixture (SM) and probabilistic mixture (PM), where the two behavioral modes are fixed and probabilistically adopted, respectively. In SM, hard conditional cooperators enhance cooperation when the threshold is low but hinder it otherwise. Surprisingly, in PM, the cooperation prevalence exhibits two first-order phase transitions as the probability is varied, leading to high, low, and vanishing levels of cooperation. Analysis of Q-tables offers insights into the "psychological shifts" of soft conditional cooperators and the overall evolutionary dynamics. Model extensions confirm the robustness of our findings. These results highlight the novel complexities arising from the diversity of conditional cooperators.

q-bio.PE↗

arXiv2025

A multi-detector neutral helium atom microscope

Scanning helium microscopy (SHeM) is an emerging technique that uses a beam of neutral atoms to image and analyse surfaces. The low energies ($\sim$64 meV) and completely non-destructive nature of the probe particles provide exceptional sensitivity for studying delicate samples and thin devices, including 2D materials. To date, around five such instruments have been constructed and are described in the literature. All represent the first attempts at SHeM construction in different laboratories, and use a single detection device. Here, we describe our second generation microscope, which is the first to offer multi-detector capabilities. The new instrument builds on recent research into SHeM optimisation and incorporates many improved design features over our previous instrument. We present measurements that highlight some of the unique capabilities the instrument provides, including 3D surface profiling, alternative imaging modes, and simultaneous acquisition of images from a mixed species beam.

physics.ins-det↗

arXiv2024

RGBDS-SLAM: A RGB-D Semantic Dense SLAM Based on 3D Multi Level Pyramid Gaussian Splatting

High-quality reconstruction is crucial for dense SLAM. Recent popular approaches utilize 3D Gaussian Splatting (3D GS) techniques for RGB, depth, and semantic reconstruction of scenes. However, these methods often overlook issues of detail and consistency in different parts of the scene. To address this, we propose RGBDS-SLAM, a RGB-D semantic dense SLAM system based on 3D multi-level pyramid gaussian splatting, which enables high-quality dense reconstruction of scene RGB, depth, and semantics.In this system, we introduce a 3D multi-level pyramid gaussian splatting method that restores scene details by extracting multi-level image pyramids for gaussian splatting training, ensuring consistency in RGB, depth, and semantic reconstructions. Additionally, we design a tightly-coupled multi-features reconstruction optimization mechanism, allowing the reconstruction accuracy of RGB, depth, and semantic maps to mutually enhance each other during the rendering optimization process. Extensive quantitative, qualitative, and ablation experiments on the Replica and ScanNet public datasets demonstrate that our proposed method outperforms current state-of-the-art methods. The open-source code will be available at: https://github.com/zhenzhongcao/RGBDS-SLAM.

cs.CV↗

arXiv2024

Navigating Spatial Inequities in Freight Truck Crash Severity via Counterfactual Inference in Los Angeles

Freight truck-related crashes pose significant challenges, leading to substantial economic losses, injuries, and fatalities, with pronounced spatial disparities across different regions. This study adopts a transport geography perspective to examine spatial justice concerns by employing deep counterfactual inference models to analyze how socioeconomic disparities, road infrastructure, and environmental conditions influence the geographical distribution and severity of freight truck crashes. By integrating road network datasets, socioeconomic attributes, and crash records from the Los Angeles metropolitan area, this research provides a nuanced spatial analysis of how different communities are disproportionately impacted. The results reveal significant spatial disparities in crash severity across areas with varying population densities, income levels, and minority populations, highlighting the pivotal role of infrastructural and environmental improvements in mitigating these disparities. The findings offer insights into targeted, location-specific policy interventions, suggesting enhancements in road infrastructure, lighting, and traffic control systems, particularly in low-income and minority-concentrated areas. This research contributes to the literature on transport geography and spatial equity by providing data-driven insights into effective measures for reducing spatial injustices associated with freight truck-related crashes.

cs.LG↗

arXiv2024

CoPS: Empowering LLM Agents with Provable Cross-Task Experience Sharing

Sequential reasoning in agent systems has been significantly advanced by large language models (LLMs), yet existing approaches face limitations. Reflection-driven reasoning relies solely on knowledge in pretrained models, limiting performance in novel scenarios, while experience-assisted reasoning often depends on external experiences and lacks clear principles for selecting representative experiences. We address these limitations by proposing CoPS (Cross-Task Experience Sharing), a generalizable algorithm that enhances sequential reasoning by cross-task experience sharing and selection. In detail, CoPS leverages agents' experiences on previous tasks, selecting distribution-matched experiences via a provable pessimism-based strategy to maximize utility while minimizing risks from distribution shifts. Extensive experimental results on benchmarks like Alfworld, Webshop, and HotPotQA demonstrate that CoPS consistently outperforms state-of-the-art baselines, with superior sample efficiency suitable for resource-constrained scenarios. Theoretically, we show that the performance of our algorithm depends on both the quality of the pretrained LLM and the matching between the agent's task-dependent trial distribution and that generated by the LLM. Our work bridges the gap between existing sequential reasoning paradigms and validates the effectiveness of leveraging cross-task experiences, shedding light on the potential to improve agents' generalization and adaptability across diverse tasks. Our codes are available at $\href{https://github.com/uclaml/COPS}{\text{https://github.com/uclaml/COPS}}$.

cs.LG↗

arXiv2024

Helium atom micro-diffraction as a characterisation tool for 2D materials

We present helium atom micro-diffraction as an ideal technique for characterization of 2D materials due to its ultimate surface sensitivity combined with sub-micron spatial resolution. Thermal energy neutral helium scatters from the valence electron density, 2-3A above the ionic cores of a surface, making the technique ideal for studying 2D materials, where other approaches can struggle due to small interaction cross-sections with few-layer samples. Sub-micron spatial resolution is key development in neutral atom scattering to allow measurements from device-scale samples. We present measurements of monolayer-substrate interactions, thermal expansion coefficients, the electron-phonon coupling constant and vacancy-type defect density on monolayer-MoS2. We also discuss extensions to the presented methods which can be immediately implemented on existing instruments to perform spatial mapping of these material properties.

physics.app-ph↗

arXiv2024

Configurable Foundation Models: Building LLMs from a Modular Perspective

Advancements in LLMs have recently unveiled challenges tied to computational efficiency and continual scalability due to their requirements of huge parameters, making the applications and evolution of these models on devices with limited computation resources and scenarios requiring various abilities increasingly cumbersome. Inspired by modularity within the human brain, there is a growing tendency to decompose LLMs into numerous functional modules, allowing for inference with part of modules and dynamic assembly of modules to tackle complex tasks, such as mixture-of-experts. To highlight the inherent efficiency and composability of the modular approach, we coin the term brick to represent each functional module, designating the modularized structure as configurable foundation models. In this paper, we offer a comprehensive overview and investigation of the construction, utilization, and limitation of configurable foundation models. We first formalize modules into emergent bricks - functional neuron partitions that emerge during the pre-training phase, and customized bricks - bricks constructed via additional post-training to improve the capabilities and knowledge of LLMs. Based on diverse functional bricks, we further present four brick-oriented operations: retrieval and routing, merging, updating, and growing. These operations allow for dynamic configuration of LLMs based on instructions to handle complex tasks. To verify our perspective, we conduct an empirical analysis on widely-used LLMs. We find that the FFN layers follow modular patterns with functional specialization of neurons and functional neuron partitions. Finally, we highlight several open issues and directions for future research. Overall, this paper aims to offer a fresh modular perspective on existing LLM research and inspire the future creation of more efficient and scalable foundational models.

cs.AI↗

arXiv2024

SELF-GUIDE: Better Task-Specific Instruction Following via Self-Synthetic Finetuning

Large language models (LLMs) hold the promise of solving diverse tasks when provided with appropriate natural language prompts. However, prompting often leads models to make predictions with lower accuracy compared to finetuning a model with ample training data. On the other hand, while finetuning LLMs on task-specific data generally improves their performance, abundant annotated datasets are not available for all tasks. Previous work has explored generating task-specific data from state-of-the-art LLMs and using this data to finetune smaller models, but this approach requires access to a language model other than the one being trained, which introduces cost, scalability challenges, and legal hurdles associated with continuously relying on more powerful LLMs. In response to these, we propose SELF-GUIDE, a multi-stage mechanism in which we synthesize task-specific input-output pairs from the student LLM, then use these input-output pairs to finetune the student LLM itself. In our empirical evaluation of the Natural Instructions V2 benchmark, we find that SELF-GUIDE improves the performance of LLM by a substantial margin. Specifically, we report an absolute improvement of approximately 15% for classification tasks and 18% for generation tasks in the benchmark's metrics. This sheds light on the promise of self-synthesized data guiding LLMs towards becoming task-specific experts without any external learning signals.

cs.CL↗

arXiv2024

TeacherLM: Teaching to Fish Rather Than Giving the Fish, Language Modeling Likewise

Large Language Models (LLMs) exhibit impressive reasoning and data augmentation capabilities in various NLP tasks. However, what about small models? In this work, we propose TeacherLM-7.1B, capable of annotating relevant fundamentals, chain of thought, and common mistakes for most NLP samples, which makes annotation more than just an answer, thus allowing other models to learn "why" instead of just "what". The TeacherLM-7.1B model achieved a zero-shot score of 52.3 on MMLU, surpassing most models with over 100B parameters. Even more remarkable is its data augmentation ability. Based on TeacherLM-7.1B, we augmented 58 NLP datasets and taught various student models with different parameters from OPT and BLOOM series in a multi-task setting. The experimental results indicate that the data augmentation provided by TeacherLM has brought significant benefits. We will release the TeacherLM series of models and augmented datasets as open-source.

cs.CL↗

arXiv2024

Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligence

The rapid advancement of large language models (LLMs) has paved the way for the development of highly capable autonomous agents. However, existing multi-agent frameworks often struggle with integrating diverse capable third-party agents due to reliance on agents defined within their own ecosystems. They also face challenges in simulating distributed environments, as most frameworks are limited to single-device setups. Furthermore, these frameworks often rely on hard-coded communication pipelines, limiting their adaptability to dynamic task requirements. Inspired by the concept of the Internet, we propose the Internet of Agents (IoA), a novel framework that addresses these limitations by providing a flexible and scalable platform for LLM-based multi-agent collaboration. IoA introduces an agent integration protocol, an instant-messaging-like architecture design, and dynamic mechanisms for agent teaming and conversation flow control. Through extensive experiments on general assistant tasks, embodied AI tasks, and retrieval-augmented generation benchmarks, we demonstrate that IoA consistently outperforms state-of-the-art baselines, showcasing its ability to facilitate effective collaboration among heterogeneous agents. IoA represents a step towards linking diverse agents in an Internet-like environment, where agents can seamlessly collaborate to achieve greater intelligence and capabilities. Our codebase has been released at \url{https://github.com/OpenBMB/IoA}.

cs.CL↗

arXiv2024

Enabling Intelligent Interactions between an Agent and an LLM: A Reinforcement Learning Approach

Large language models (LLMs) encode a vast amount of world knowledge acquired from massive text datasets. Recent studies have demonstrated that LLMs can assist an embodied agent in solving complex sequential decision making tasks by providing high-level instructions. However, interactions with LLMs can be time-consuming. In many practical scenarios, it requires a significant amount of storage space that can only be deployed on remote cloud servers. Additionally, using commercial LLMs can be costly since they may charge based on usage frequency. In this paper, we explore how to enable intelligent cost-effective interactions between a down stream task oriented agent and an LLM. We find that this problem can be naturally formulated by a Markov decision process (MDP), and propose When2Ask, a reinforcement learning based approach that learns when it is necessary to query LLMs for high-level instructions to accomplish a target task. On one side, When2Ask discourages unnecessary redundant interactions, while on the other side, it enables the agent to identify and follow useful instructions from the LLM. This enables the agent to halt an ongoing plan and transition to a more suitable one based on new environmental observations. Experiments on MiniGrid and Habitat environments that entail planning sub-goals demonstrate that When2Ask learns to solve target tasks with only a few necessary interactions with the LLM, significantly reducing interaction costs in testing environments compared with baseline methods. Our code is available at: https://github.com/ZJLAB-AMMI/LLM4RL.

cs.AI↗

arXiv2024

MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies

The burgeoning interest in developing Large Language Models (LLMs) with up to trillion parameters has been met with concerns regarding resource efficiency and practical expense, particularly given the immense cost of experimentation. This scenario underscores the importance of exploring the potential of Small Language Models (SLMs) as a resource-efficient alternative. In this context, we introduce MiniCPM, specifically the 1.2B and 2.4B non-embedding parameter variants, not only excel in their respective categories but also demonstrate capabilities on par with 7B-13B LLMs. While focusing on SLMs, our approach exhibits scalability in both model and data dimensions for future LLM research. Regarding model scaling, we employ extensive model wind tunnel experiments for stable and optimal scaling. For data scaling, we introduce a Warmup-Stable-Decay (WSD) learning rate scheduler (LRS), conducive to continuous training and domain adaptation. We present an in-depth analysis of the intriguing training dynamics that occurred in the WSD LRS. With WSD LRS, we are now able to efficiently study data-model scaling law without extensive retraining experiments on both axes of model and data, from which we derive the much higher compute optimal data-model ratio than Chinchilla Optimal. Additionally, we introduce MiniCPM family, including MiniCPM-DPO, MiniCPM-MoE and MiniCPM-128K, whose excellent performance further cementing MiniCPM's foundation in diverse SLM applications. MiniCPM models are available publicly at https://github.com/OpenBMB/MiniCPM .

cs.CL↗

arXiv2024

Large Language Model as a Policy Teacher for Training Reinforcement Learning Agents

Recent studies have uncovered the potential of Large Language Models (LLMs) in addressing complex sequential decision-making tasks through the provision of high-level instructions. However, LLM-based agents lack specialization in tackling specific target problems, particularly in real-time dynamic environments. Additionally, deploying an LLM-based agent in practical scenarios can be both costly and time-consuming. On the other hand, reinforcement learning (RL) approaches train agents that specialize in the target task but often suffer from low sampling efficiency and high exploration costs. In this paper, we introduce a novel framework that addresses these challenges by training a smaller, specialized student RL agent using instructions from an LLM-based teacher agent. By incorporating the guidance from the teacher agent, the student agent can distill the prior knowledge of the LLM into its own model. Consequently, the student agent can be trained with significantly less data. Moreover, through further training with environment feedback, the student agent surpasses the capabilities of its teacher for completing the target task. We conducted experiments on challenging MiniGrid and Habitat environments, specifically designed for embodied AI research, to evaluate the effectiveness of our framework. The results clearly demonstrate that our approach achieves superior performance compared to strong baseline methods. Our code is available at https://github.com/ZJLAB-AMMI/LLM4Teach.

cs.AI↗

arXiv2024

Microcavity induced by few-layer GaSe crystal on silicon photonic crystal waveguide for efficient optical frequency conversion

We demonstrate the post-induction of high-quality microcavity on silicon photonic crystal (PC) waveguide by integrating few-layer GaSe crystal, which promises highly efficient on-chip optical frequency conversions. The integration of GaSe shifts the dispersion bands of the PC waveguide mode into the bandgap, resulting in localized modes confined by the bare PC waveguides. Thanks to the small contrast of refractive index at the boundaries of microcavity, it is reliably to obtain quality (Q) factors exceeding 10^4. With the enhanced light-GaSe interaction by the microcavity modes and high second-order nonlinearity of GaSe, remarkable second-harmonic generation (SHG) and sum-frequency generation (SFG) are achieved. A record-high on-chip SHG conversion efficiency of 131100% W^-1 is obtained, enabling the clear SHG imaging of the resonant modes with the pump of sub-milliwatts continuous-wave (CW) laser. Driven by a pump of on-resonance CW laser, strong SFGs are successfully carried out with the other pump of a CW laser spanning over the broad telecom-band. Broadband frequency conversion of an incoherent superluminescent light-emitting diode with low spectral power density is also realized in the integrated GaSe-PC waveguide. Our results are expected to provide new strategies for high-efficiency light-matter interactions, nonlinear photonics and light source generation in silicon photonic integrated circuits.

physics.optics↗

arXiv2024

Emergence of cooperation under punishment: A reinforcement learning perspective

Punishment is a common tactic to sustain cooperation and has been extensively studied for a long time. While most of previous game-theoretic work adopt the imitation learning where players imitate the strategies who are better off, the learning logic in the real world is often much more complex. In this work, we turn to the reinforcement learning paradigm, where individuals make their decisions based upon their past experience and long-term returns. Specifically, we investigate the Prisoners' dilemma game with Q-learning algorithm, and cooperators probabilistically pose punishment on defectors in their neighborhood. Interestingly, we find that punishment could lead to either continuous or discontinuous cooperation phase transitions, and the nucleation process of cooperation clusters is reminiscent of the liquid-gas transition. The uncovered first-order phase transition indicates that great care needs to be taken when implementing the punishment compared to the continuous scenario.

q-bio.PE↗