arXiv subjects

Yuan Yang

Publications and source records attributed to Yuan Yang.

At least 37 records · Page 2Linked to original sources

arXiv2025

GatePro: Parameter-Free Expert Selection Optimization for Mixture-of-Experts Models

Modern large language models leverage Mixture-of-Experts (MoE) architectures for efficient scaling, but face a critical challenge: functionally similar experts are often selected simultaneously, creating redundant computation and limiting effective model capacity. Existing auxiliary balance loss methods improve token distribution but fail to address the underlying expert diversity problem. We introduce GatePro, a novel parameter-free method that directly promotes expert selection diversity. GatePro identifies the most similar expert pairs and introduces localized competition mechanisms, preventing redundant expert co-activation while maintaining natural expert specialization. Our comprehensive evaluation demonstrates GatePro's effectiveness across model scales and benchmarks. Analysis demonstrates GatePro's ability to achieve enhanced expert diversity, where experts develop more distinct and complementary capabilities, avoiding functional redundancy. This approach can be deployed hot-swappable during any training phase without additional learnable parameters, offering a practical solution for improving MoE effectiveness.

cs.CL↗

arXiv2025

Image-Based Visual Servoing for Enhanced Cooperation of Dual-Arm Manipulation

The cooperation of a pair of robot manipulators is required to manipulate a target object without any fixtures. The conventional control methods coordinate the end-effector pose of each manipulator with that of the other using their kinematics and joint coordinate measurements. Yet, the manipulators' inaccurate kinematics and joint coordinate measurements can cause significant pose synchronization errors in practice. This paper thus proposes an image-based visual servoing approach for enhancing the cooperation of a dual-arm manipulation system. On top of the classical control, the visual servoing controller lets each manipulator use its carried camera to measure the image features of the other's marker and adapt its end-effector pose with the counterpart on the move. Because visual measurements are robust to kinematic errors, the proposed control can reduce the end-effector pose synchronization errors and the fluctuations of the interaction forces of the pair of manipulators on the move. Theoretical analyses have rigorously proven the stability of the closed-loop system. Comparative experiments on real robots have substantiated the effectiveness of the proposed control.

cs.RO↗

arXiv2025

Balanced Actor Initialization: Stable RLHF Training of Distillation-Based Reasoning Models

The development of alignment and reasoning capabilities in large language models has seen remarkable progress through two paradigms: instruction tuning and reinforcement learning from human feedback (RLHF) alignment paradigm, and distillation-based reasoning fine-tuning paradigm. While both approaches prove effective independently, the third paradigm of applying RLHF to distillation-trained models presents significant challenges. Our investigation reveals two critical phenomena that emerge in this paradigm: Sequence Length Collapse, where language generation dramatically reduces during early RLHF training, and the Reward Hockey Stick Curve, featuring severe reward score drops followed by gradual recovery. These instabilities fundamentally compromise the model's alignment and reasoning capabilities. To address these challenges, we propose Balanced Actor Initialization (BAI), a two-stage weighted model merging approach. BAI first merges instruction-following and distillation-based reasoning fine-tuned models, then further combines this intermediate model with the pretrained model to preserve foundational knowledge. Through comprehensive experiments across diverse benchmarks and detailed analysis of training experiments, we demonstrate that BAI resolves Sequence Length Collapse, mitigates the Reward Hockey Stick Curve, and enables continuous sequence length improvement during training. Additionally, our analysis reveals that balanced merging ratios achieve optimal trade-offs between training stability and reasoning capability preservation. Our work provides the effective solution for stable training in this third paradigm, enabling more capable reasoning models that combine distillation efficiency with RLHF alignment.

cs.CL↗

arXiv2025

Deliberate Reasoning in Language Models as Structure-Aware Planning with an Accurate World Model

Enhancing the reasoning capabilities of language models (LMs) remains a key challenge, especially for tasks that require complex, multi-step decision-making where existing Chain-of-Thought (CoT) approaches struggle with consistency and verification. In this paper, we propose a novel reasoning framework, referred to as Structure-aware Planning with an Accurate World Model (SWAP), that integrates structured knowledge representation with learned planning. Unlike prior methods that rely purely on natural language reasoning, SWAP leverages entailment graphs to encode structured dependencies and enable symbolic verification of intermediate steps. To systematically construct and update the graph, SWAP employs a policy model to propose candidate expansions and a world model to predict structural updates. To improve accuracy, the world model generates multiple alternative updates, and a discriminator re-ranks them based on plausibility. To encourage diverse exploration, we introduce Diversity-based Modelling (DM), which samples candidates from the remaining probability mass after removing previously sampled candidates from the original policy distribution. Additionally, SWAP improves the discrimination accuracy through Contrastive Ranking (CR), which directly compares candidates within prompts and incorporates meta-knowledge to improve ranking quality. We evaluate SWAP across diverse reasoning-intensive benchmarks including math reasoning, logical reasoning, and coding tasks. Extensive experiments demonstrate that SWAP significantly improves upon the base models and consistently outperforms existing reasoning methods.

cs.CL↗

arXiv2025

Evolution from intralayer to interlayer superconductivity in a bilayer $t$-$J$ model

Motivated by the bilayer cuprate superconductors and nickelate superconductor La$_3$Ni$_2$O$_7$, we investigate the evolution from intralayer to interlayer superconductivity based on a bilayer two-leg $t$-$J$-$J_{\bot}$ model, where $t$ is the in-plane electron hopping, $J$ is the in-plane spin interaction, and $J_{\bot}$ is the inter-plane spin interaction. By means of the density matrix renormalization group calculations, we obtain the quantum phase diagram of the system by tuning $J_{\bot}$ in a large doping range $δ= 1/8 - 1/2$. We find that a large $J_{\bot}$ can always drive an interlayer superconductivity by coupling the two layers in both the Luther-Emery liquid and Luttinger liquid states. By coupling two Luther-Emery liquid states, the in-plane superconductivity evolves to inter-plane superconductivity either through an intermediate charge density wave (CDW) phase or directly, depending on doping ratio. This emergent CDW phase, which exists over a finite doping range, appears to develop from the CDW state of the two-leg ladder at $δ= 1/4$. By coupling two Luttinger liquids, the in-plane Luttinger liquids show a transition to the inter-plane superconducting phase at large $J_{\bot}$, as reported in previous literature. Interestingly, in the intermediate $J_{\bot}$ regime we find that while the in-plane Luttinger-liquid features remain stable, the inter-plane superconductivity can develop an enhanced quasi-long-range order with the power exponent $K^{zz}_{\rm SC} \sim 1$. At last, we show that the interlayer superconductivity is also stable by coupling the bilayer three-leg $t$-$J$ ladders by a strong $J_{\bot}$ interaction, from both the Luther-Emery liquid and Luttinger-liquid states.

cond-mat.str-el↗

arXiv2025

Two Is Better Than One: Rotations Scale LoRAs

Scaling Low-Rank Adaptation (LoRA)-based Mixture-of-Experts (MoE) facilitates large language models (LLMs) to efficiently adapt to diverse tasks. However, traditional gating mechanisms that route inputs to the best experts may fundamentally hinder LLMs' scalability, leading to poor generalization and underfitting issues. We identify that the root cause lies in the restricted expressiveness of existing weighted-sum mechanisms, both within and outside the convex cone of LoRA representations. This motivates us to propose RadarGate, a novel geometrically inspired gating method that introduces rotational operations of LoRAs representations to boost the expressiveness and facilitate richer feature interactions among multiple LoRAs for scalable LLMs. Specifically, we first fuse each LoRA representation to other LoRAs using a learnable component and then feed the output to a rotation matrix. This matrix involves learnable parameters that define the relative angular relationship between LoRA representations. Such a simple yet effective mechanism provides an extra degree of freedom, facilitating the learning of cross-LoRA synergies and properly tracking the challenging poor generalization and underfitting issues as the number of LoRA grows. Extensive experiments on 6 public benchmarks across 21 tasks show the effectiveness of our RadarGate for scaling LoRAs. We also provide valuable insights, revealing that the rotations to each pair of representations are contrastive, encouraging closer alignment of semantically similar representations during geometrical transformation while pushing distance ones further apart. We will release our code to the community.

cs.LG↗

arXiv2025

Distinct hydrologic response patterns and trends worldwide revealed by physics-embedded learning

To track rapid changes within our water sector, Global Water Models (GWMs) need to realistically represent hydrologic systems' response patterns - such as baseflow fraction - but are hindered by their limited ability to learn from data. Here we introduce a high-resolution physics-embedded big-data-trained model as a breakthrough in reliably capturing characteristic hydrologic response patterns ('signatures') and their shifts. By realistically representing the long-term water balance, the model revealed widespread shifts - up to ~20% over 20 years - in fundamental green-blue-water partitioning and baseflow ratios worldwide. Shifts in these response patterns, previously considered static, contributed to increasing flood risks in northern mid-latitudes, heightening water supply stresses in southern subtropical regions, and declining freshwater inputs to many European estuaries, all with ecological implications. With more accurate simulations at monthly and daily scales than current operational systems, this next-generation model resolves large, nonlinear seasonal runoff responses to rainfall ('elasticity') and streamflow flashiness in semi-arid and arid regions. These metrics highlight regions with management challenges due to large water supply variability and high climate sensitivity, but also provide tools to forecast seasonal water availability. This capability newly enables global-scale models to deliver reliable and locally relevant insights for water management.

physics.geo-ph↗

arXiv2025

TILP: Differentiable Learning of Temporal Logical Rules on Knowledge Graphs

Compared with static knowledge graphs, temporal knowledge graphs (tKG), which can capture the evolution and change of information over time, are more realistic and general. However, due to the complexity that the notion of time introduces to the learning of the rules, an accurate graph reasoning, e.g., predicting new links between entities, is still a difficult problem. In this paper, we propose TILP, a differentiable framework for temporal logical rules learning. By designing a constrained random walk mechanism and the introduction of temporal operators, we ensure the efficiency of our model. We present temporal features modeling in tKG, e.g., recurrence, temporal order, interval between pair of relations, and duration, and incorporate it into our learning process. We compare TILP with state-of-the-art methods on two benchmark datasets. We show that our proposed framework can improve upon the performance of baseline methods while providing interpretable results. In particular, we consider various scenarios in which training samples are limited, data is biased, and the time range between training and inference are different. In all these cases, TILP works much better than the state-of-the-art methods.

cs.CL↗

arXiv2024

The Compressor-Retriever Architecture for Language Model OS

Recent advancements in large language models (LLMs) have significantly enhanced their capacity to aggregate and process information across multiple modalities, enabling them to perform a wide range of tasks such as multimodal data querying, tool usage, web interactions, and handling long documents. These capabilities pave the way for transforming LLMs from mere chatbots into general-purpose agents capable of interacting with the real world. This paper explores the concept of using a language model as the core component of an operating system (OS), effectively acting as a CPU that processes data stored in a context window, which functions as RAM. A key challenge in realizing such an LM OS is managing the life-long context and ensuring statefulness across sessions, a feature limited by the current session-based interaction paradigm due to context window size limit. To address this, we introduce compressor-retriever, a model-agnostic architecture designed for life-long context management. Unlike other long-context solutions such as retrieval-augmented generation, our approach exclusively uses the base model's forward function to compress and retrieve context, ensuring end-to-end differentiability. Preliminary experiments demonstrate the effectiveness of this architecture in in-context learning tasks, marking a step towards the development of a fully stateful LLM OS. Project repo available at: https://github.com/gblackout/LM-OS

cs.CL↗

arXiv2024

Can LLMs Reason in the Wild with Programs?

Large Language Models (LLMs) have shown superior capability to solve reasoning problems with programs. While being a promising direction, most of such frameworks are trained and evaluated in settings with a prior knowledge of task requirements. However, as LLMs become more capable, it is necessary to assess their reasoning abilities in more realistic scenarios where many real-world problems are open-ended with ambiguous scope, and often require multiple formalisms to solve. To investigate this, we introduce the task of reasoning in the wild, where an LLM is tasked to solve a reasoning problem of unknown type by identifying the subproblems and their corresponding formalisms, and writing a program to solve each subproblem, guided by a tactic. We create a large tactic-guided trajectory dataset containing detailed solutions to a diverse set of reasoning problems, ranging from well-defined single-form reasoning (e.g., math, logic), to ambiguous and hybrid ones (e.g., commonsense, combined math and logic). This allows us to test various aspects of LLMs reasoning at the fine-grained level such as the selection and execution of tactics, and the tendency to take undesired shortcuts. In experiments, we highlight that existing LLMs fail significantly on problems with ambiguous and mixed scope, revealing critical limitations and overfitting issues (e.g. accuracy on GSM8K drops by at least 50\%). We further show the potential of finetuning a local LLM on the tactic-guided trajectories in achieving better performance. Project repo is available at github.com/gblackout/Reason-in-the-Wild

cs.CL↗

arXiv2024

Alterations of electrocortical activity during hand movements induced by motor cortex glioma

Glioma cells can reshape functional neuronal networks by hijacking neuronal synapses, leading to partial or complete neurological dysfunction. These mechanisms have been previously explored for language functions. However, the impact of glioma on sensorimotor functions is still unknown. Therefore, we recruited a control group of patients with unaffected motor cortex and a group of patients with glioma-infiltrated motor cortex, and recorded high-density electrocortical signals during finger movement tasks. The results showed that glioma suppresses task-related synchronization in the high-gamma band and reduces the power across all frequency bands. The resulting atypical motor information transmission model with discrete signaling pathways and delayed responses disrupts the stability of neuronal encoding patterns for finger movement kinematics across various temporal-spatial scales. These findings demonstrate that gliomas functionally invade neural circuits within the motor cortex. This result advances our understanding of motor function processing in chronic disease states, which is important to advance the surgical strategies and neurorehabilitation approaches for patients with malignant gliomas.

q-bio.NC↗

arXiv2024

Temporal Inductive Logic Reasoning over Hypergraphs

Inductive logic reasoning is a fundamental task in graph analysis, which aims to generalize patterns from data. This task has been extensively studied for traditional graph representations, such as knowledge graphs (KGs), using techniques like inductive logic programming (ILP). Existing ILP methods assume learning from KGs with static facts and binary relations. Beyond KGs, graph structures are widely present in other applications such as procedural instructions, scene graphs, and program executions. While ILP is beneficial for these applications, applying it to those graphs is nontrivial: they are more complex than KGs, which usually involve timestamps and n-ary relations, effectively a type of hypergraph with temporal events. In this work, we propose temporal inductive logic reasoning (TILR), an ILP method that reasons on temporal hypergraphs. To enable hypergraph reasoning, we introduce the multi-start random B-walk, a novel graph traversal method for hypergraphs. By combining it with a path-consistency algorithm, TILR learns logic rules by generalizing from both temporal and relational data. To address the lack of hypergraph benchmarks, we create and release two temporal hypergraph datasets: YouCook2-HG and nuScenes-HG. Experiments on these benchmarks demonstrate that TILR achieves superior reasoning capability over various strong baselines.

cs.LG↗

arXiv2024

TEILP: Time Prediction over Knowledge Graphs via Logical Reasoning

Conventional embedding-based models approach event time prediction in temporal knowledge graphs (TKGs) as a ranking problem. However, they often fall short in capturing essential temporal relationships such as order and distance. In this paper, we propose TEILP, a logical reasoning framework that naturally integrates such temporal elements into knowledge graph predictions. We first convert TKGs into a temporal event knowledge graph (TEKG) which has a more explicit representation of time in term of nodes of the graph. The TEKG equips us to develop a differentiable random walk approach to time prediction. Finally, we introduce conditional probability density functions, associated with the logical rules involving the query interval, using which we arrive at the time prediction. We compare TEILP with state-of-the-art methods on five benchmark datasets. We show that our model achieves a significant improvement over baselines while providing interpretable explanations. In particular, we consider several scenarios where training samples are limited, event types are imbalanced, and forecasting the time of future events based on only past events is desired. In all these cases, TEILP outperforms state-of-the-art methods in terms of robustness.

cs.CL↗

arXiv2023

A Neurodiversity-Inspired Solver for the Abstraction \& Reasoning Corpus (ARC) Using Visual Imagery and Program Synthesis

Core knowledge about physical objects -- e.g., their permanency, spatial transformations, and interactions -- is one of the most fundamental building blocks of biological intelligence across humans and non-human animals. While AI techniques in certain domains (e.g. vision, NLP) have advanced dramatically in recent years, no current AI systems can yet match human abilities in flexibly applying core knowledge to solve novel tasks. We propose a new AI approach to core knowledge that combines 1) visual representations of core knowledge inspired by human mental imagery abilities, especially as observed in studies of neurodivergent individuals; with 2) tree-search-based program synthesis for flexibly combining core knowledge to form new reasoning strategies on the fly. We demonstrate our system's performance on the very difficult Abstraction \& Reasoning Corpus (ARC) challenge, and we share experimental results from publicly available ARC items as well as from our 4th-place finish on the private test set during the 2022 global ARCathon challenge.

cs.AI↗

arXiv2023

A Cognitively-Inspired Neural Architecture for Visual Abstract Reasoning Using Contrastive Perceptual and Conceptual Processing

We introduce a new neural architecture for solving visual abstract reasoning tasks inspired by human cognition, specifically by observations that human abstract reasoning often interleaves perceptual and conceptual processing as part of a flexible, iterative, and dynamic cognitive process. Inspired by this principle, our architecture models visual abstract reasoning as an iterative, self-contrasting learning process that pursues consistency between perceptual and conceptual processing of visual stimuli. We explain how this new Contrastive Perceptual-Conceptual Network (CPCNet) works using matrix reasoning problems in the style of the well-known Raven's Progressive Matrices intelligence test. Experiments on the machine learning dataset RAVEN show that CPCNet achieves higher accuracy than all previously published models while also using the weakest inductive bias. We also point out a substantial and previously unremarked class imbalance in the original RAVEN dataset, and we propose a new variant of RAVEN -- AB-RAVEN -- that is more balanced in terms of abstract concepts.

cs.AI↗

arXiv2023

FERN: Leveraging Graph Attention Networks for Failure Evaluation and Robust Network Design

Robust network design, which aims to guarantee network availability under various failure scenarios while optimizing performance/cost objectives, has received significant attention. Existing approaches often rely on model-based mixed-integer optimization that is hard to scale or employ deep learning to solve specific engineering problems yet with limited generalizability. In this paper, we show that failure evaluation provides a common kernel to improve the tractability and scalability of existing solutions. By providing a neural network function approximation of this common kernel using graph attention networks, we develop a unified learning-based framework, FERN, for scalable Failure Evaluation and Robust Network design. FERN represents rich problem inputs as a graph and captures both local and global views by attentively performing feature extraction from the graph. It enables a broad range of robust network design problems, including robust network validation, network upgrade optimization, and fault-tolerant traffic engineering that are discussed in this paper, to be recasted with respect to the common kernel and thus computed efficiently using neural networks and over a small set of critical failure scenarios. Extensive experiments on real-world network topologies show that FERN can efficiently and accurately identify key failure scenarios for both OSPF and optimal routing scheme, and generalizes well to different topologies and input traffic patterns. It can speed up multiple robust network design problems by more than 80x, 200x, 10x, respectively with negligible performance gap.

cs.NI↗

arXiv2023

A Computational Account Of Self-Supervised Visual Learning From Egocentric Object Play

Research in child development has shown that embodied experience handling physical objects contributes to many cognitive abilities, including visual learning. One characteristic of such experience is that the learner sees the same object from several different viewpoints. In this paper, we study how learning signals that equate different viewpoints -- e.g., assigning similar representations to different views of a single object -- can support robust visual learning. We use the Toybox dataset, which contains egocentric videos of humans manipulating different objects, and conduct experiments using a computer vision framework for self-supervised contrastive learning. We find that representations learned by equating different physical viewpoints of an object benefit downstream image classification accuracy. Further experiments show that this performance improvement is robust to variations in the gaps between viewpoints, and that the benefits transfer to several different image classification tasks.

cs.CV↗

arXiv2023

Harnessing the Power of Large Language Models for Natural Language to First-Order Logic Translation

Translating natural language sentences to first-order logic (NL-FOL translation) is a longstanding challenge in the NLP and formal logic literature. This paper introduces LogicLLaMA, a LLaMA-7B model fine-tuned for NL-FOL translation using LoRA on a single GPU. LogicLLaMA is capable of directly translating natural language into FOL rules, which outperforms GPT-3.5. LogicLLaMA is also equipped to correct FOL rules predicted by GPT-3.5, and can achieve similar performance as GPT-4 with a fraction of the cost. This correction ability was achieved by a novel supervised fine-tuning (SFT) + reinforcement learning with human feedback (RLHF) framework, which initially trains on synthetically perturbed NL-FOL pairs to encourage chain-of-thought reasoning and then fine-tunes with RLHF on GPT-3.5 outputs using a FOL verifier as the reward model. To train LogicLLaMA, we present MALLS (large language $\textbf{M}$odel gener$\textbf{A}$ted N$\textbf{L}$-FO$\textbf{L}$ pair$\textbf{S}$), a dataset of 34K high-quality and diverse sentence-level NL-FOL pairs collected from GPT-4. The dataset was created by implementing a pipeline that prompts GPT-4 for pairs, and dynamically adjusts the prompts to ensure the collection of pairs with rich and diverse contexts at different levels of complexity, and verifies the validity of the generated FOL rules. Codes, weights, and data are available at $\href{https://github.com/gblackout/LogicLLaMA}{\small \text{https://github.com/gblackout/LogicLLaMA}}$.

cs.CL↗