arXiv subjects

Liyun Zhang

Publications and source records attributed to Liyun Zhang.

At least 19 recordsLinked to original sources

arXiv2026

AgentDiff: Meaning-Bearing Rewrites Trigger Deeper Divergence than Presentation Changes in LLM Agents

LLM agents should respond to what an input means, not how it is presented. We show that they do not treat these two kinds of variation equally. AgentDiff measures the difference between answer changes caused by meaning-bearing rewrites (paraphrase and synonym substitution) and presentation changes (reordering, formatting, and distractors), while matching perturbation severity. Across 68 model--benchmark--scaffold cells spanning ten LLMs from seven architecture families, three benchmarks, 1{,}530 original questions, and approximately 11{,}150 variants, meaning-bearing rewrites produce a $+19.69$ percentage-point higher inconsistency rate (paired $t=9.58$, $p<0.0001$; 64/68 cells positive). The result is stable under four severity proxies ($+18.9$ to $+20.9$~pp, all $p<0.0001$) and remains $+11.10$~pp on the 48 cells outside the qwen family. A fully held-out qwen2.5-14B-Instruct evaluation then tests the structure of this effect on 1{,}800 new trajectories: the pre-registered capable-and-tractable partition is positive in 3/4 held-out cells and remains sharply separated after pooling (Welch $t=3.81$, $p=9.6\!\times\!10^{-4}$). Trace analysis explains how the difference propagates. Meaning-bearing rewrites preserve the first action but reduce thought similarity from step 2 onward by $5.6$--$10.5$ points and extend the resulting cascade by $0.17$ steps (paired $t=7.69$, $p=2.5\!\times\!10^{-14}$), a pattern we call \emph{stealth divergence}. AgentDiff therefore establishes a reproducible directional robustness gap, identifies the regime in which it is most reliable, and connects final-answer inconsistency to a distinct trajectory-level signature. Code, perturbations, trajectories, and analysis scripts are released for review at https://anonymous.4open.science/r/agentdiff-emnlp-0BB4/

cs.CL↗

arXiv2026

Heteroscedastic Neural Surrogate Modeling for Robust and Rapid Bayesian Inference in Fusion Plasma Diagnostics

Bayesian inference via Markov Chain Monte Carlo (MCMC) provides effective parameter estimation, but its real-time application in complex physical systems is hindered by heavy computational bottlenecks and extreme sensitivity to statistical noise. We address this by proposing a neural-network-based probabilistic surrogate framework for rapid and robust MCMC inference. Using fusion plasma Thomson scattering diagnostics as a challenging, noise-dominated testbed, our approach employs a dual-head architecture to simultaneously estimate the expected physical emission spectrum and the channel-wise intrinsic measurement noise variance. By optimizing a Gaussian Negative Log-Likelihood (GNLL) objective, the learned aleatoric uncertainty dynamically buffers the sampler against pathological shot noise. Evaluations demonstrate that this surrogate framework achieves > 1500x acceleration over exact physical forward models, while simultaneously reducing inference error (RMSE) by >20% compared to standard homoscedastic neural baselines, offering a highly promising paradigm for real-time physical analysis.

stat.CO↗

arXiv2026

ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models

Vision-Language-Action (VLA) models exhibit remarkable action generation for embodied intelligence, but their heavy compute make deployment on edge platforms impractical. Aggressive, sub-4-bit weight quantization is the natural solution, yet existing post-training quantization (PTQ) methods suffer severe performance degradation in this regime. To address this, we introduce ActQuant, an action-guided mixed-precision PTQ framework that operates in two stages: (1) an inter-tensor bit allocator that assigns each weight matrix a single bit-width based on how much it contributes to predicting the agent's actions; (2) an intra-tensor scale optimizer tunes per-block quantization scales using action-aware curvature, so that dynamic range is concentrated on the weights most influential for control. To deliver the on-device benefits of our aggressive quantization, we further introduce OmniModel.cpp, an agentic conversion pipeline that ports architectures into a native C/C++ runtime with efficient low-bit kernels. We evaluate ActQuant both in simulation and on a real-world 6-DoF UR3 arm, with all models deployed through OmniModel.cpp. On the LIBERO benchmark, ActQuant is the only method that operates at or below 3 bits-per-weight, retaining 95.0% on OpenVLA-OFT and 94.8% on $π_{0.5}$. Pushed further, ActQuant reaches 2.5 bpw at 90.1% on OpenVLA-OFT, compressing the backbone from 14.3 GB to 2.7 GB (5.3$\times$). On the physical UR3 arm, $π_{0.5}$ quantized with ActQuant retains the baseline's success rate while reducing the memory footprint by 2.5$\times$.

cs.CV↗

arXiv2026

A Pilot Study of Mildly Recycled Pulsars: A Case Study of PSR J2338+4818

Mildly recycled pulsars are neutron stars partially spun up through relatively short mass-transfer phases, typically with massive carbon-oxygen (CO) or oxygen-neon-magnesium (ONeMg) white dwarf companions. PSR J2338+4818, a mildly recycled pulsar, was discovered with the Five-hundred-meter Aperture Spherical Telescope (FAST). As a pilot study on the formation and evolutionary pathways of mildly recycled pulsars, we present the updated timing solution for PSR J2338+4818 and examine its single pulses and scintillation properties. Aided by the sensitivity of FAST, the single pulses of PSR J2338+4818 were systematically studied. 27,228 single pulses with S/N > 7 have been detected in our observations. For the FAST ultra-wideband observation on MJD 61045, the receiver was still in the technical commissioning phase, and then only a preliminary single-pulse search was performed. Pulse nulling was examined using a Markov Chain Monte Carlo (MCMC) method, but no evidence for nulling was found. The possible long-term nulling reported by previous studies did not occur in any of our observations in either the 1.0 to 1.5 GHz band or the 300 to 600 MHz band. Interstellar scintillation is evident in our observations. The measured scintillation timescales and bandwidths range from 2.93 to 25.26 minutes and 1.68 to 27.41 MHz, respectively. In all observations, no clear scintillation arc was found in the secondary spectra of PSR J2338+4818.

astro-ph.HE↗

arXiv2026

Isolated neutron star candidates from the fourth generation XMM-Newton catalogues

X-ray thermally emitting isolated neutron stars (XINSs) are a rare population that provides insights into neutron star cooling, magnetic-field evolution, and Galactic demographics. Using more than two decades of observations from the European Space Agency's XMM-Newton Observatory, we searched the 4XMM-DR9 and 4XMM-DR12 catalogues for absorbed XINS candidates down to a flux of $10^{-14}$ erg cm$^{-2}$ s$^{-1}$ in the 0.5--1 keV band. Candidates were selected based on soft X-ray spectra and the absence of catalogued optical, ultraviolet, or infrared counterparts. Follow-up observations with XMM-Newton and FAST were complemented by data from the SRG/eROSITA All-Sky Survey, Chandra, and optical surveys. Of ten sources analysed, five are compelling XINS candidates, one is the known XINS 4XMM J022141.5-735632, two are extragalactic contaminants, and two remain ambiguous because of limited photon statistics. The five candidates exhibit soft ($kT\sim80-100$ eV), moderately absorbed, and stable X-ray emission consistent with distant XINSs. They are located primarily in the Galactic plane, with possible associations at distances of $\sim$1.8 and $\sim$6 kpc. Population-synthesis simulations predict $20\pm5$ XINSs within the 4XMM-DR12 footprint, of which $6^{+2}_{-3}$ exceed our flux threshold, consistent with the observed sample if additional candidates are confirmed. The model further predicts that $\sim$70% of the population remains below the detection threshold. Deep optical and additional X-ray observations are required to establish the nature of the candidates. Future missions such as NewAthena will enable more detailed studies of these distant populations and improve constraints on the Galactic XINS population.

astro-ph.HE↗

arXiv2026

Event-Driven Proactive Assistive Manipulation with Grounded Vision-Language Planning

Assistance in collaborative manipulation is often initiated by user instructions, making high-level reasoning request-driven. In fluent human teamwork, however, partners often infer the next helpful step from the observed outcome of an action rather than waiting for instructions. Motivated by this, we introduce a shift from request-driven assistance to event-driven proactive assistance, where robot actions are initiated by workspace state transitions induced by human--object interactions rather than user-provided task instructions. To this end, we propose an event-driven framework that tracks interaction progress with an event monitor and, upon event completion, extracts stabilized pre/post snapshots that characterize the resulting state transition. Given the stabilized snapshots, the planner analyzes the implied state transition to infer a task-level goal and decide whether to intervene; if so, it generates a sequence of assistive actions. To make outputs executable and verifiable, we restrict actions to a set of action primitives and reference objects via integer IDs. We evaluate the framework on a real tabletop number-block collaboration task, demonstrating that explicit pre/post state-change evidence improves proactive completion on solvable scenes and appropriate waiting on unsolvable ones.

cs.RO↗

arXiv2026

AcoustEmo: Open-Vocabulary Emotion Reasoning via Utterance-Aware Acoustic Q-Former

Multimodal Large Language Models (MLLMs) excel in Open-Vocabulary (OV) emotion recognition but often neglect fine-grained acoustic modeling. Existing methods typically use global audio encoders, failing to capture subtle, local temporal dynamics like micro-prosody and intonation shifts within individual utterances. To address this, we propose AcoustEmo, a time-sensitive MLLM featuring a novel Utterance-Aware Acoustic Q-Former. Our approach utilizes a timestamp-synchronized sliding window to dynamically extract segment-level audio tokens instead of coarse global representations. This enables the model to explicitly trace the temporal evolution of subtle acoustic clues and capture deep contextual dependencies in dialogues. Experiments on the Explainable Multimodal Emotion Recognition (EMER) task show that AcoustEmo significantly enhances complex emotion reasoning, outperforming baselines while maintaining robust contextual accuracy.

cs.MM↗

arXiv2026

From emission to absorption: the FAST observation of the OH 18-cm lines from the Comet C2025/A6

We observed comet C/2025 A6 with the Five-hundred-meter Aperture Spherical radio Telescope (FAST) equipped with the ultra-wideband receiver from 2025 October 23 to November 8, and it was the first detection for this comet with FAST.Through trapezoidal fitting of the OH line profiles, we derived the expansion velocities of the water which showed an increase from 1.5$\pm$0.3 km s$^{-1}$ at the heliocentric distance of 0.65 AU to 3.0$\pm$0.9 km s$^{-1}$ at 0.54 AU. Based on these results, we estimated the OH production rates of C/2025 A6 for October 23, October 26, November 4 and November 5 which were (1.0$\pm$0.1)$\times$10$^{29}$, (1.2$\pm$0.1)$\times$10$^{29}$, (1.4$\pm$0.3)$\times$10$^{29}$, and (1.5$\pm$0.4)$\times$10$^{29}$ s$^{-1}$ respectively. The results show a significant upward trend.

astro-ph.EP↗

arXiv2026

Spectral Dataset of Stripped-Envelope Supernovae from the Tsinghua Supernova Group

The extent of envelope stripping in the progenitor stars is directly reflected in the diversity of spectral features observed in stripped-envelope supernovae (SESNe). Through extensive spectral observation and analysis, we aim to clarify the statistical differences between the subclasses of SESNe. The Tsinghua Supernova group obtained 249 optical spectra of 62 SESNe during the years from 2010 to 2020, covering phases from $-$16 to over 190 days relative to maximum light. Most spectra were obtained during the photospheric phases after the supernova explosion. For each spectrum, the pseudo-equivalent widths (pEWs) and blueshift velocities of principal lines were measured. We further investigated the common spectral features by analysing their velocity and strength correlations across all subtypes. We identify the feature near 6200~Å in SNe Ib as H$\mathrmα$ through comparison with SNe IIb and Ic, which resolves inconsistent literature interpretations. Our finding reveals prevalent residual hydrogen in SNe Ib, further supporting a continuous stripping sequence from SNe IIb to Ib. We observe a trend in increasing velocity among different subtypes of stripped-envelope SNe, with SNe IIb exhibiting the lowest line velocities, followed by Ib, Ic, and Ic-BL. Typically, the O~I lines in SNe Ic/Ic-BL are stronger than those seen in SNe IIb/Ib. In nebular phases, the [Ca II] emission dominates over [O I] in SNe IIb/Ib while [O I] is stronger in SNe Ic, including the He-rich SN 2016coi. This spectral dichotomy implies that progenitors of SNe Ic (BL) have more massive CO cores and hence higher initial masses.

astro-ph.HE↗

arXiv2026

Search for Periodic Radio Signals from Double Neutron Star System Companions Using the Fast Folding Algorithm

As most of the companions in the double neutron star systems should be normal pulsars, the Fast Folding Algorithm (FFA), which is suitable for finding these long spin period pulsars, was used to search their possible radio signals. A time domain resampling code PYSOLATOR was used to maximize the available data length by removing the orbital modulation. We collected and processed 272.2 hours observational data taken by the Five-hundred-meter Aperture Spherical radio Telescope (FAST) for the 13 double neutron star systems in its sky. The signal-to-noise ratios of known pulsar signals are obviously improved by this search method, including the detection of a faint pulsar signal which only saw by folding the data. Unfortunately, no companion signals were found among all the 197962 candidates. Geodetic precession of the orbit could enhance detectability in future observations.

astro-ph.HE↗

arXiv2026

A Unified Evaluation Framework for Multi-Annotator Tendency Learning

Recent works have emerged in multi-annotator learning that shift focus from Consensus-oriented Learning (CoL), which aggregates multiple annotations into a single ground-truth prediction, to Individual Tendency Learning (ITL), which models annotator-specific labeling behavior patterns (i.e., tendency) to provide explanation analysis for understanding annotator decisions. However, no evaluation framework currently exists to assess whether ITL methods truly capture individual tendencies and provide meaningful behavioral explanations. To address this gap, we propose the first unified evaluation framework with two novel metrics: (1) Difference of Inter-annotator Consistency (DIC) quantifies how well models capture annotator tendencies by comparing predicted inter-annotator similarity structures with ground-truth; (2) Behavior Alignment Explainability (BAE) evaluates how well model explanations reflect annotator behavior and decision relevance by aligning explainability-derived with ground-truth labeling similarity structures via Multidimensional Scaling (MDS). Extensive experiments validate the effectiveness of our proposed evaluation framework.

cs.LG↗

arXiv2026

3DGesPolicy: Phoneme-Aware Holistic Co-Speech Gesture Generation Based on Action Control

Generating holistic co-speech gestures that integrate full-body motion with facial expressions suffers from semantically incoherent coordination on body motion and spatially unstable meaningless movements due to existing part-decomposed or frame-level regression methods, We introduce 3DGesPolicy, a novel action-based framework that reformulates holistic gesture generation as a continuous trajectory control problem through diffusion policy from robotics. By modeling frame-to-frame variations as unified holistic actions, our method effectively learns inter-frame holistic gesture motion patterns and ensures both spatially and semantically coherent movement trajectories that adhere to realistic motion manifolds. To further bridge the gap in expressive alignment, we propose a Gesture-Audio-Phoneme (GAP) fusion module that can deeply integrate and refine multi-modal signals, ensuring structured and fine-grained alignment between speech semantics, body motion, and facial expressions. Extensive quantitative and qualitative experiments on the BEAT2 dataset demonstrate the effectiveness of our 3DGesPolicy across other state-of-the-art methods in generating natural, expressive, and highly speech-aligned holistic gestures.

cs.CV↗

arXiv2025

Push the Limit of Multi-modal Emotion Recognition by Prompting LLMs with Receptive-Field-Aware Attention Weighting

Understanding the emotions in a dialogue usually requires external knowledge to accurately understand the contents. As the LLMs become more and more powerful, we do not want to settle on the limited ability of the pre-trained language model. However, the LLMs either can only process text modality or are too expensive to process the multimedia information. We aim to utilize both the power of LLMs and the supplementary features from the multimedia modalities. In this paper, we present a framework, Lantern, that can improve the performance of a certain vanilla model by prompting large language models with receptive-field-aware attention weighting. This framework trained a multi-task vanilla model to produce probabilities of emotion classes and dimension scores. These predictions are fed into the LLMs as references to adjust the predicted probabilities of each emotion class with its external knowledge and contextual understanding. We slice the dialogue into different receptive fields, and each sample is included in exactly t receptive fields. Finally, the predictions of LLMs are merged with a receptive-field-aware attention-driven weighting module. In the experiments, vanilla models CORECT and SDT are deployed in Lantern with GPT-4 or Llama-3.1-405B. The experiments in IEMOCAP with 4-way and 6-way settings demonstrated that the Lantern can significantly improve the performance of current vanilla models by up to 1.23% and 1.80%.

cs.CL↗

arXiv2025

3DFacePolicy: Audio-Driven 3D Facial Animation Based on Action Control

Audio-driven 3D facial animation has achieved significant progress in both research and applications. While recent baselines struggle to generate natural and continuous facial movements due to their frame-by-frame vertex generation approach, we propose 3DFacePolicy, a pioneer work that introduces a novel definition of vertex trajectory changes across consecutive frames through the concept of "action". By predicting action sequences for each vertex that encode frame-to-frame movements, we reformulate vertex generation approach into an action-based control paradigm. Specifically, we leverage a robotic control mechanism, diffusion policy, to predict action sequences conditioned on both audio and vertex states. Extensive experiments on VOCASET and BIWI datasets demonstrate that our approach significantly outperforms state-of-the-art methods and is particularly expert in dynamic, expressive and naturally smooth facial animations.

cs.CV↗

arXiv2025

Millisecond Pulsars in M2: New discoveries and a detailed timing analysis

Globular clusters (GCs) offer a unique environment for discovering and studying millisecond pulsars. In this paper, we present a multi-epoch search and detailed timing analysis of millisecond pulsars in the GC M2, using the Five-hundred-meter Aperture Spherical Telescope. We have discovered two new binary millisecond pulsars in M2, designated M2F and M2G, respectively. We provide measurements of the emission properties of all known pulsars in M2, including their polarization profiles, rotation measures, flux densities, scintillation characteristics, and so forth. In particular, we report the first rotation measure at the distance and direction of this cluster. Additionally, we report the first phase-coherent timing solutions for the M2 pulsars. From our Bayesian timing analysis, we have measured their spin and orbital parameters with high precision, including the advance of periastron for M2A and M2E indicating total system masses of 1.75(13) and 1.80(5) solar masses respectively. Using archival data from the Hubble Space Telescope, we have identified an optical counterpart of M2C, which is likely the white dwarf companion of the pulsar. By combining results from optical and radio observations, we have reconstructed the binary evolution track of this system and estimated the cooling age of the companion to be approximately 10\,Myr, making it the youngest white dwarf in any known GC binary pulsars. Furthermore, using the spin period derivatives of M2 pulsars, we have investigated the gravitational potential of the cluster and found that our results strongly support the latest central-stellar-velocity dispersion measurement in M2.

astro-ph.HE↗

arXiv2025

SimLabel: Similarity-Weighted Iterative Framework for Multi-annotator Learning with Missing Annotations

Multi-annotator learning (MAL) aims to model annotator-specific labeling patterns. However, existing methods face a critical challenge: they simply skip updating annotator-specific model parameters when encountering missing labels, i.e., a common scenario in real-world crowdsourced datasets where each annotator labels only small subsets of samples. This leads to inefficient data utilization and overfitting risks. To this end, we propose a novel similarity-weighted semi-supervised learning framework (SimLabel) that leverages inter-annotator similarities to generate weighted soft labels for missing annotations, enabling the utilization of unannotated samples rather than skipping them entirely. We further introduce a confidence-based iterative refinement mechanism that combines maximum probability with entropy-based uncertainty to prioritize predicted high-quality pseudo-labels to impute missing labels, jointly enhancing similarity estimation and model performance over time. For evaluation, we contribute a new multimodal multi-annotator dataset, AMER2, with high and more variable missing rates, reflecting real-world annotation sparsity and enabling evaluation across different sparsity levels.

cs.MM↗

arXiv2025

QuMAB: Query-based Multi-Annotator Behavior Modeling with Reliability under Sparse Labels

Multi-annotator learning traditionally aggregates diverse annotations to approximate a single ground truth, treating disagreements as noise. However, this paradigm faces fundamental challenges: subjective tasks often lack absolute ground truth, and sparse annotation coverage makes aggregation statistically unreliable. We introduce a paradigm shift from sample-wise aggregation to annotator-wise behavior modeling. By treating annotator disagreements as valuable information rather than noise, modeling annotator-specific behavior patterns can reconstruct unlabeled data to reduce annotation cost, enhance aggregation reliability, and explain annotator decision behavior. To this end, we propose QuMAB (Query-based Multi-Annotator Behavior Pattern Learning), which uses light-weight queries to model individual annotators while capturing inter-annotator correlations as implicit regularization, preventing overfitting to sparse individual data while maintaining individualization and improving generalization, with a visualization of annotator focus regions offering an explainable analysis of behavior understanding. We contribute two large-scale datasets with dense per-annotator labels: STREET (4,300 labels/annotator) and AMER (average 3,118 labels/annotator), the first multimodal multi-annotator dataset. Extensive experiments demonstrate the superiority of our QuMAB in modeling individual annotators' behavior patterns, their utility for consensus prediction, and applicability under sparse annotations.

cs.MM↗

arXiv2025

REPRO-Bench: Can Agentic AI Systems Assess the Reproducibility of Social Science Research?

Assessing the reproducibility of social science papers is essential for promoting rigor in research processes, but manual assessment is costly. With recent advances in agentic AI systems (i.e., AI agents), we seek to evaluate their capability to automate this process. However, existing benchmarks for reproducing research papers (1) focus solely on reproducing results using provided code and data without assessing their consistency with the paper, (2) oversimplify real-world scenarios, and (3) lack necessary diversity in data formats and programming languages. To address these issues, we introduce REPRO-Bench, a collection of 112 task instances, each representing a social science paper with a publicly available reproduction report. The agents are tasked with assessing the reproducibility of the paper based on the original paper PDF and the corresponding reproduction package. REPRO-Bench features end-to-end evaluation tasks on the reproducibility of social science papers with complexity comparable to real-world assessments. We evaluate three representative AI agents on REPRO-Bench, with the best-performing agent achieving an accuracy of only 21.4%. Building on our empirical analysis, we develop REPRO-Agent, which improves the highest accuracy achieved by existing agents by 71%. We conclude that more advanced AI agents should be developed to automate real-world reproducibility assessment. REPRO-Bench is publicly available at https://github.com/uiuc-kang-lab/REPRO-Bench.

cs.CL↗