Searcharxiv⌕ Search

arXiv subjects

Yutong Chen

Publications and source records attributed to Yutong Chen.

At least 19 recordsLinked to original sources
arXiv2026

MV-dVRK: A Multi-Viewpoint Benchmark for Spatial Surgical Perception

Large-scale training and refined optimization techniques have greatly improved sparse multi-view 3D reconstruction. Despite their relevance to surgery, such methods have never before been rigorously evaluated on real endoscopic images. Current clinical telerobots deploy a single stereo camera inside the patient, making multi-viewpoint data extremely rare. This paper presents MV-dVRK, the first ex-vivo surgical dataset to combine multiple exposure-synchronized stereo viewpoints with accurate surface geometry and camera poses. The static subset of the benchmark provides dense SfM reference geometry, validated against an industrial 3D scanner, together with ground-truth camera poses and sparse-view test sets. We use MV-dVRK to systematically compare zero-shot monocular, stereo, multi-stereo, and multi-view 3D reconstruction methods as the number of viewpoints increases. With two endoscopes, multi-stereo reconstruction achieves the highest coverage. With a third viewpoint, optimization-based multi-view methods perform best, covering 67% of ground-truth surface points within a 1 mm tolerance and recovering highly accurate relative camera poses. By contrast, feed-forward foundation models cover only 43% of the ground-truth surface in the same setting. MV-dVRK also includes ten dynamic sequences spanning multiple surgical tasks, with increasing kinematic complexity and tissue deformation, providing a basis for future research in multi-viewpoint surgical perception. The project is available at: https://mv-dvrk.is.mpg.de.

cs.CV↗
arXiv2026

DualOPSD: Adaptive Privileged Teachers for On-Policy Self-Distillation

On-policy self-distillation (OPSD) uses a privileged copy of the student model to provide dense supervision without an external teacher. OPSD keeps this privileged teacher fixed, even though the student distribution and output style change during training. We propose DualOPSD, an asymmetric alternating framework that adapts both policies. The student first learns from the privileged teacher. The teacher then moves toward the updated student distribution on the same student trajectory. This update makes later supervision responsive to the learner and does not require another rollout. On Qwen3-8B in non-thinking mode, DualOPSD improves avg@12 over OPSD by 23.61, 13.89, and 10.00 points on AIME 2024, AIME 2025, and HMMT 2025. Results at 1.7B and 4B show that the accuracy gain depends on model scale. Across all three scales, DualOPSD reduces truncation. The 4B diagnostic also shows lower KL in both directions between the teacher and student.

cs.LG↗
arXiv2026

PACEShop: Evaluating Personalized, Actionable, Compositional, and Evidence-grounded Shopping Assistants

Shopping assistants are shifting from ranked product lists toward structured decision support, where systems must synthesize shopper context, product evidence, and next-step guidance into a coherent recommendation experience. This changes the unit of evaluation: a fluent response can still fail by ignoring shopper context, contradicting itself across components, or leaving defects too vague to localize. Existing personalization, grounding, and LLM-as-a-judge benchmarks cover pieces of this problem, but they do not define a joint evaluation target for structured shopping-assistant responses. We formulate this missing evaluation target as PACE: Personalized, Actionable, Compositional, and Evidence-grounded evaluation. We instantiate PACE with two artifacts: PACEShop, a benchmark dataset that makes the target measurable through 22,625 controlled records with structured personas, auditable evidence pools, GOOD/BAD labels, and gold defect family and location annotations; and PACEJudge, a training-free judging protocol that makes the target reportable through a structured output contract. Our experiments show that generic judges can recognize broad quality but fail to recover the diagnostic fields required for PACE; PACEShop makes these failures verifiable, and PACEJudge improves persona-source, cross-component, grounding, and family/location closure without retraining, showing that realistic shopping-assistant evaluation requires a task-matched output contract rather than only a stronger backbone or scalar prompt.

cs.CL↗
arXiv2026

Penelope: Localized Latent Recurrence for Efficient Structured Reasoning

Complex structured reasoning tasks often require additional computation, yet current language models obtain it mainly by increasing parameter scale or by serializing intermediate steps as chain-of-thought (CoT) tokens. The former raises training and deployment costs, while the latter ties reasoning computation to autoregressive output length. We introduce Penelope, an efficient latent-reasoning framework for pretrained decoder-only Transformers that localizes recurrent computation to a selected decoder interval. The lower decoder prefix is evaluated once to construct a problem-conditioned boundary memory, which is then iteratively refined through time-modulated GRU dynamics and recurrent readout states before answer generation. A progressive CoT-to-latent curriculum transfers visible reasoning into this internal recurrent path, allowing additional computation to be allocated in latent space without repeatedly executing the complete decoder or generating a long intermediate trace. Experiments on open-source structured-reasoning benchmarks show that, at validation-selected latent budgets, Penelope attains competitive accuracy relative to established latent-reasoning models while reducing measured inference latency. These results show that latent refinement can be localized to a narrow decoder interval, reducing repeated full-decoder execution without generating a long visible reasoning trace and providing a practical accuracy-efficiency tradeoff for decoder-only Transformer models.

cs.AI↗
arXiv2026

TraceCLIP: Recovering Local Semantics from Patch-to-CLS Contributions

Dense vision-language understanding, including object localization, region recognition, and open-vocabulary semantic segmentation, requires associating language concepts with spatially grounded visual regions. CLIP provides a strong foundation for these tasks by learning a shared image-text embedding space from large-scale contrastive pre-training. However, its image-level objective aligns text with a CLS-derived global representation, leaving local vision-language correspondence only indirectly constrained. Existing methods either introduce additional supervision, external models, or task-specific adaptation, while training-free approaches mainly recover dense responses from existing patch features without examining where local semantics become most accessible within CLIP. We introduce TraceCLIP, a training-free framework that recovers latent patch-level semantic evidence by isolating the patch-specific terms written into the CLS attention output. TraceCLIP further converts contribution-derived semantic responses into a semantic-geodesic topology gate that calibrates final-layer patch affinity for dense feature reconstruction. Diagnostic experiments show that these contribution features exhibit strong local semantic discrimination and text-conditioned spatial alignment. On eight zero-shot semantic segmentation benchmarks, TraceCLIP achieves gains of 1.3 to 4.5 points in average mIoU over the strongest prior training-free methods across both backbones and background settings, without additional training, external vision foundation models, or region-level supervision. More broadly, these findings suggest that spatially localized semantics may remain accessible within the internal construction of globally aligned representations.

cs.CV↗
arXiv2026

A Disconnected Superconducting Regime at the Parent Limit of Infinite-Layer Nickelates

Infinite-layer nickelates have been widely viewed as cuprate analogs in which superconductivity emerges and forms a superconducting dome centered around 10-20% cation substitution. Here we show that pristine and stoichiometric PrNiO2, without cation substitution, exhibits intrinsic superconductivity characterized by zero resistance and diamagnetism in uncapped films. Through heterostructure engineering, we further exclude an interfacial origin of the superconductivity. Remarkably, zero-resistance superconductivity is consistently observed in trivalent-substituted PrNiO2, whereas it is rapidly suppressed by dilute divalent substitution. Combined with angle-resolved photoemission studies, these results indicate that such a new superconducting regime is confined to within 3% additional hole doping from pristine PrNiO2. Furthermore, this phase is separated from the previously established superconducting dome around ~ 20% divalent doping by a non-superconducting region in the phase diagram, and is further distinguished by a remarkably stronger upper-critical-field anisotropy. These findings establish a unique separated superconducting regime, suggesting that infinite-layer nickelates are not merely cuprate analogs but host distinct superconducting physics.

cond-mat.supr-con↗
arXiv2026

Marginal-Fermi-Liquid-like Behavior without Pseudogap in Infinite-Layer Nickelates

Pseudogap formation, strange-metal behavior and unconventional superconductivity are closely intertwined in hole-doped cuprates, yet their relationship remains unresolved. Infinite-layer nickelates offer a distinct 3d9-derived platform to address this question by combining a cuprate-like Ni dx2-y2 Fermi surface with multiband electronic degrees of freedom. Here we use angle-resolved photoemission spectroscopy to resolve the low-energy spectral function of superconducting La0.8Ca0.2NiO2 and parent LaNiO2 thin films. In La0.8Ca0.2NiO2, the electronic self-energy Im Sigma(omega) is approximately linear in energy and its slope increases from (pi/2, pi/2) to (pi, 0), revealing momentum-dependent marginal-Fermi-liquid-like scattering. Both films show a progressive suppression of low-energy spectral weight from the diagonal direction toward (pi, 0), with stronger suppression in parent LaNiO2. However, finite Fermi-level spectral weight persists around the entire Fermi surface, with no leading-edge shift or back-bending indicative of pseudogap formation in either the electron pocket or the cuprate-like hole band. Our results demonstrate that momentum-selective correlations and marginal-Fermi-liquid-like scattering can occur without a detectable cuprate-like pseudogap, providing a benchmark for identifying the essential normal-state electronic ingredients of high-temperature superconductivity.

cond-mat.supr-con↗
arXiv2026

GGPT: Geometry Grounded Point Transformer

Recent feed-forward networks have achieved remarkable progress in sparse-view 3D reconstruction by predicting dense point maps directly from RGB images. However, they often suffer from geometric inconsistencies and limited fine-grained accuracy due to the absence of explicit multi-view constraints. We introduce the Geometry-Grounded Point Transformer (GGPT), a framework that augments feed-forward reconstruction with reliable sparse geometric guidance. We first propose an improved Structure-from-Motion pipeline based on dense feature matching and lightweight geometric optimisation to efficiently estimate accurate camera poses and partial 3D point clouds from sparse input views. Building on this foundation, we propose a geometry-guided 3D point transformer that refines dense point maps under explicit partial-geometry supervision using an optimised guidance encoding. Extensive experiments demonstrate that our method provides a principled mechanism for integrating geometric priors with dense feed-forward predictions, producing reconstructions that are both geometrically consistent and spatially complete, recovering fine structures and filling gaps in textureless areas. Trained solely on ScanNet++ with VGGT predictions, GGPT generalises across architectures and datasets, substantially outperforming state-of-the-art feed-forward 3D reconstruction models in both in-domain and out-of-domain settings.

cs.CV↗
arXiv2026

TimeRadar: A Domain-Rotatable Foundation Model for Time Series Anomaly Detection

Current time series foundation models (TSFMs) primarily focus on learning prevalent and regular patterns within a predefined time or frequency domain to enable supervised downstream tasks (e.g., forecasting). Consequently, they are often ineffective for inherently unsupervised downstream tasks-such as time series anomaly detection (TSAD), which aims to identify rare, irregular patterns. This limitation arises because such abnormal patterns can closely resemble the regular patterns when presented in the same time/frequency domain. To address this issue, we introduce TimeRadar, an innovative TSFM built in a fractional time-frequency domain to support generalist TSAD across diverse unseen datasets. Our key insight is that rotating a time series into a data-dependent fractional time-frequency representation can adaptively differentiate the normal and abnormal signals across different datasets. To this end, a novel component, namely Fractionally modulated Time-Frequency Reconstruction (FTFRecon), is proposed in TimeRadar to leverage a learnable fractional order to rotate the time series to the most pronounced angle between a continuous time and frequency domain for accurate data reconstruction. This provides adaptive data reconstruction in an optimal time-frequency domain for each data input, enabling effective differentiation of the unbounded abnormal patterns from the regular ones across datasets, including unseen datasets. To allow TimeRadar to model local abnormality that is not captured by the global data reconstruction, we further introduce a Contextual Deviation Learning (CDL) component to model the local deviation of the input relative to its contextual time series data in the rotatable domain.

cs.LG↗
arXiv2026

Collaborative Parameter Learning: Mitigating Forgetting via Parameter-Level Gradient Analysis

Catastrophic forgetting during knowledge injection impairs the ability of large language models to acquire new knowledge without overwriting previously mastered knowledge. Recent studies analyze forgetting from a gradient similarity perspective and mitigate forgetting through vector projection. However, these methods primarily characterize gradient similarity at the aggregate direction level, leaving the parameter wise contributions to forgetting underexplored. In this paper, we decompose gradient similarity into parameter wise contributions and identify two types of parameters during forgetting: Conflicting Parameters, whose updates contribute to forgetting and typically account for 50 percent to 75 percent of parameters, and Collaborative Parameters, whose updates mitigate forgetting and account for 25 percent to 50 percent. Based on this analysis, we propose Collaborative Parameter Learning, CPL, a parameter wise training rule that freezes Conflicting Parameters and updates only Collaborative Parameters. Experiments comparing CPL with seven baseline methods show that CPL learns 20.2% to 48.2% more questions with negligible forgetting, while reducing peak VRAM by approximately 3 GB per billion model parameters and computation time by 16.5 percent. Extensive evaluations on parameter consumption, out of set generalization, cross prompt generalization, multimodal tasks, open ended question answering, and multilingual settings demonstrate that CPL effectively mitigates forgetting across diverse scenarios.

cs.LG↗
arXiv2026

CoScale-RL: Efficient Post-Training by Co-Scaling Data and Computation

Training Large Reasoning Model (LRM) is usually unstable and unpredictable, especially on hard problems or weak foundation models. We found that the current post-training scaling strategy can still improve on these cases. We propose CoScale-RL, a novel scaling strategy with better data and computational efficiency. We first scale up solutions to make problems solvable. The core idea is to collect multiple solutions for each problem, rather than simply enlarging the dataset. Then, we scale up rollout computation to stabilize Reinforcement Learning. We further leverage a model merge technique called Re-distillation to sustain or even improve computational efficiency when scaling up. Our method significantly improves data and computational efficiency, with an average 3.76$\times$ accuracy improvement on four benchmarks. CoScale-RL is able to improve an LRM's ability boundary without an extensive SFT dataset. Our method provides a new scaling direction to further improve LRM's reasoning ability.

cs.LG↗
arXiv2026

Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline

Recent advances in LLM-based multi-agent systems (MAS) show that workflows composed of multiple LLM agents with distinct roles, tools, and communication patterns can outperform single-LLM baselines on complex tasks. However, most frameworks are homogeneous, where all agents share the same base LLM and differ only in prompts, tools, and positions in the workflow. This raises the question of whether such workflows can be simulated by a single agent through multi-turn conversations. We investigate this across seven benchmarks spanning coding, mathematics, general question answering, domain-specific reasoning, and real-world planning and tool use. Our results show that a single agent can reach the performance of homogeneous workflows with an efficiency advantage from KV cache reuse, and can even match the performance of an automatically optimized heterogeneous workflow. Building on this finding, we propose \textbf{OneFlow}, an algorithm that automatically tailors workflows for single-agent execution, reducing inference costs compared to existing automatic multi-agent design frameworks without trading off accuracy. These results position the single-LLM implementation of multi-agent workflows as a strong baseline for MAS research. We also note that single-LLM methods cannot capture heterogeneous workflows due to the lack of KV cache sharing across different LLMs, highlighting future opportunities in developing \textit{truly} heterogeneous multi-agent systems.

cs.MA↗
arXiv2026

ENTRA: Entropy-Based Redundancy Avoidance in Large Language Model Reasoning

Large Reasoning Models (LRMs) often suffer from overthinking, generating unnecessarily long reasoning chains even for simple tasks. This leads to substantial computational overhead with limited performance gain, primarily due to redundant verification and repetitive generation. While prior work typically constrains output length or optimizes correctness, such coarse supervision fails to guide models toward concise yet accurate inference. In this paper, we propose ENTRA, an entropy-based training framework that suppresses redundant reasoning while preserving performance. ENTRA first estimates the token-level importance using a lightweight Bidirectional Importance Estimation (BIE) method, which accounts for both prediction confidence and forward influence. It then computes a redundancy reward based on the entropy of low-importance tokens, normalized by its theoretical upper bound, and optimizes this reward via reinforcement learning. Experiments on mathematical reasoning benchmarks demonstrate that ENTRA reduces output length by 37% to 53% with no loss-and in some cases, gains-in accuracy. Our approach offers a principled and efficient solution to reduce overthinking in LRMs, and provides a generalizable path toward redundancy-aware reasoning optimization.

cs.AI↗
arXiv2025

Towards Effective Model Editing for LLM Personalization

Personalization is becoming indispensable for LLMs to align with individual user preferences and needs. Yet current approaches are often computationally expensive, data-intensive, susceptible to catastrophic forgetting, and prone to performance degradation in multi-turn interactions or when handling implicit queries. To address these challenges, we conceptualize personalization as a model editing task and introduce Personalization Editing, a framework that applies localized edits guided by clustered preference representations. This design enables precise preference-aligned updates while preserving overall model capabilities. In addition, existing personalization benchmarks frequently rely on persona-based dialogs between LLMs rather than user-LLM interactions, or focus primarily on stylistic imitation while neglecting information-seeking tasks that require accurate recall of user-specific preferences. We introduce User Preference Question Answering (UPQA), a short-answer QA dataset constructed from in-situ user queries with varying levels of difficulty. Unlike prior benchmarks, UPQA directly evaluates a model's ability to recall and apply specific user preferences. Across experimental settings, Personalization Editing achieves higher editing accuracy and greater computational efficiency than fine-tuning, while outperforming prompting-based baselines in multi-turn conversations and implicit preference questions settings.

cs.CL↗
arXiv2025

Nonholonomic Narrow Dead-End Escape with Deep Reinforcement Learning

Nonholonomic constraints restrict feasible velocities without reducing configuration-space dimension, which makes collision-free geometric paths generally non-executable for car-like robots. Ackermann steering further imposes curvature bounds and forbids in-place rotation, so escaping from narrow dead ends typically requires tightly sequenced forward and reverse maneuvers. Classical planners that decouple global search and local steering struggle in these settings because narrow passages occupy low-measure regions and nonholonomic reachability shrinks the set of valid connections, which degrades sampling efficiency and increases sensitivity to clearances. We study nonholonomic narrow dead-end escape for Ackermann vehicles and contribute three components. First, we construct a generator that samples multi-phase forward-reverse trajectories compatible with Ackermann kinematics and inflates their envelopes to synthesize families of narrow dead ends that are guaranteed to admit at least one feasible escape. Second, we construct a training environment that enforces kinematic constraints and train a policy using the soft actor-critic algorithm. Third, we evaluate against representative classical planners that combine global search with nonholonomic steering. Across parameterized dead-end families, the learned policy solves a larger fraction of instances, reduces maneuver count, and maintains comparable path length and planning time while under the same sensing and control limits. We provide our project as open source at https://github.com/gitagitty/cisDRL-RobotNav.git

cs.RO↗
arXiv2025

LacAIDes: Generative AI-Supported Creative Interactive Circuits Crafting to Enliven Traditional Lacquerware

Lacquerware, a representative craft of Chinese intangible cultural heritage, is renowned for its layered aesthetics and durability but faces declining engagement. While prior human-computer interaction research has explored embedding interactive circuits to transform lacquerware into responsive artifacts, most studies have focused on fabrication techniques rather than supporting makers in creatively designing such interactions at a low threshold. To address this gap, we present LacAIDes, a Generative AI powered creativity-support tool built on a multi-agent workflow aligned with the double diamond model of design thinking. LacAIDes enables exploration and creation of culturally grounded interactive circuits without requiring prior technical expertise. We evaluated LacAIDes in a longitudinal workshop with 34 participants using a mixed-method approach. Results show that LacAIDes demonstrated high usability, enhanced creative engagement in craft making, and encouraged critical reflection on the role of Generative AI in digital craft practices. This work contributes to human-computer interaction by introducing a novel creativity-support tool and providing empirical insights into revitalizing traditional craft making through Generative AI.

cs.HC↗
arXiv2025

Joint Frequency-Space Sparse Reconstruction for DOA Estimation under Coherent Sources and Amplitude-Phase Errors

In this letter, we propose a joint frequency-space sparse reconstruction method for direction-of-arrival (DOA) estimation, which effectively addresses the issues arising from the existence of coherent sources and array amplitude-phase errors. Specifically, by using an auxiliary source with known angles, we first construct the real steering vectors (RSVs) based on the spectral peaks of received signals in the frequency domain, which serve as a complete basis matrix for compensation for amplitude-phase errors. Then, we leverage the spectral sparsity of snapshot data in the frequency domain and the spatial sparsity of incident directions to perform the DOA estimation according to the sparse reconstruction method. The proposed method does not require iterative optimization, hence exhibiting low computational complexity. Numerical results demonstrate that the proposed DOA estimation method achieves higher estimation accuracy for coherent sources as compared to various benchmark schemes.

eess.SP↗
arXiv2025

Observation of Electride-like $s$ States Coexisting with Correlated $d$ Electrons in NdNiO$_2$

Despite exhibiting a similar $d_{x^2-y^2}$ band character to cuprates, infinite-layer nickelates host additional electron pockets that distinguish them from single-band cuprates. The elusive orbital origin of these electron pockets has led to competing theoretical scenarios. Here, using polarization-dependent and resonant angle-resolved photoemission spectroscopy (ARPES), we determine the orbital character of the Fermi surfaces in NdNiO$_2$. Our data reveal that the electron-like pocket arises predominantly from interstitial $s$ states, with negligible contributions from rare-earth 5$d$ and 4$f$ orbitals near the Fermi level. The observation of well-defined quantum well states indicates a uniform distribution of these interstitial electrons throughout the film thickness. By comparing with electronic structure of LaNiO$_2$, we find that the rare-earth element modulates the Ni-derived bands and hopping integrals through a chemical pressure effect. These findings clarify the role of rare-earth elements in shaping the low-energy electronic structure and establish the presence of electride-like interstitial $s$ states in a correlated oxide system, where electrons occupy lattice voids rather than atomic orbitals. The electride-like character offer new insight into the self-doping and superconductivity in infinite-layer nickelates.

cond-mat.supr-con↗