arXiv subjects

Tao Yang

Publications and source records attributed to Tao Yang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Sound-Horizon-Independent Test of Cosmic Distance Duality Relation Using Artificial Neural Networks and Gaussian Processes

The cosmic distance duality relation (CDDR), $D_L(z)(1+z)^{-2}/D_A(z) \equiv 1$, is a fundamental relation in modern cosmology linking luminosity distance $D_L$ and angular diameter distance $D_A$. We define $η(z) \equiv D_L(z)(1+z)^{-2}/D_A(z)$ and reconstruct the relevant quantities over $0 < z < 2.5$ using two independent non-parametric methods: Gaussian processes (GP) and artificial neural networks (ANN). Specifically, we reconstruct $D_L$ from three different Type Ia supernovae (SNe$~$Ia) compilations (PantheonPlus, Union3, and DES-Dovekie), the baryon acoustic oscillation distance ratio $D_M/D_H$ from SDSS and DESI, and the Hubble function $H(z)$ from cosmic chronometer (CC) data. Our CDDR test is independent of both the cosmological model and the sound horizon $r_d$ calibration, as $D_M/D_H$ naturally eliminates $r_d$. We render our GP reconstructions insensitive to mean-function and kernel choices via full Bayesian marginalization over hyperparameters, with ANN as a cross-check$-$revealing the impact of data sparsity on the latter. Our results show no significant deviation from the CDDR exceeding $2σ$, and for the CC+BAO+DES-Dovekie combination it remains within $1σ$ over most of the redshift range. With $r_d$ eliminated, varying the only remaining external parameter $M_B$ systematically shifts $η(z)$ through the $M_B$-$H_0$ degeneracy, indicating that the observed $η\neq1$ signal is closely tied to $H_0$ tension among datasets and is likely driven by $M_B$ systematics rather than a true CDDR violation. Future high-quality CC, BAO, and SNe$~$Ia data$-$with either improved $M_B$ calibration or $M_B$-free distance measurements$-$will be essential for extending this calibration-free approach.

astro-ph.CO↗

arXiv2026

LoRA Fine-Tuned Models for Control Systems Course Q\&A: A Multidimensional Evaluation of Model Scale and Rank Effects

Large language models (LLMs) are increasingly used in specialized university courses, but control-systems questions require coordinated terminology, notation, derivations, and stepwise explanations. Direct general-purpose responses may be inconsistently structured and hard to verify. Using exercises and reference solutions from a Linear Control Systems course, we built a supervised fine-tuning dataset of 360 system-user-assistant conversations. We applied LoRA to Qwen2.5-3B-Instruct and Qwen2.5-7B-Instruct. With identical data splits, inference settings, and evaluation protocols, we compared base and fine-tuned models and tested LoRA ranks r=4, 8, and 16. Evaluation used ROUGE, BERTScore, and structured-output features to measure reference-answer similarity and stability of the Solution-Method-Teaching Points format. LoRA improved both similarity and structured-output stability at both sizes. On the current test set, 7B-r16 achieved the highest ROUGE-L (0.4093) and BERTScore-F1 (0.8643), while r=8 offered a better balance between performance and parameter efficiency. Bootstrap resampling showed ROUGE-L gains of 0.0764 [0.0613, 0.0915] for 3B-r16 and 0.0874 [0.0687, 0.1042] for 7B-r16; both intervals exceeded zero, indicating stable textual-similarity improvements on the current test set. These results suggest LoRA can align open-source instruction-tuned models more closely with the language and pedagogical organization of course reference answers. However, the metrics mainly capture textual similarity and formatting consistency, not domain-specific reasoning or mathematical correctness, which require expert assessment and task-specific rubrics.

cs.AI↗

arXiv2026

Systematic Biases in Gravitational-Wave Parameter Estimation from Neglecting Orbital Eccentricity in Space-Based Detectors

Accurate modeling of gravitational-wave signals is essential for reliable inference of compact-binary source parameters, particularly for future space-based detectors operating in the milli- and deci-Hertz bands. In this work, we systematically investigate the parameter-estimation biases induced by neglecting orbital eccentricity when analyzing eccentric compact-binary coalescences with quasicircular waveform templates. Focusing on the deci-Hertz detector B-DECIGO and the milli-Hertz detector LISA, we model eccentric inspiral signals using a frequency-domain waveform that incorporates eccentricity-induced higher harmonics and the time-dependent response of spaceborne detectors. We quantify systematic biases in the chirp mass, symmetric mass ratio, and luminosity distance using both Bayesian inference and the Fisher-Cutler-Vallisneri (FCV) formalism, and assess their significance relative to statistical uncertainties. By constructing mock gravitational-wave catalogs spanning stellar-mass and massive black-hole binaries, we identify critical initial eccentricities at which systematic errors become comparable to statistical errors. We find that, for B-DECIGO, adopting a one-year observation period with f_min$ = 0.1$ Hz, even very small eccentricities, $e_0 \sim 10^{-4}-10^{-3}$ at 0.1 Hz, can lead to significant biases, whereas for LISA, adopting a five-year observation period with f_min$ = 10^{-4}$ Hz, such effects typically arise at larger eccentricities, $e_0 \sim 10^{-2}-10^{-1}$ at $10^{-4}$ Hz, due to the smaller number of in-band cycles. Comparisons between FCV predictions and full Bayesian analyses show good agreement over most of the initial eccentricity range, but diverge at high eccentricity, where Bayesian inference is required. Our results highlight the necessity of incorporating eccentricity in waveform models for future space-based gravitational-wave observations.

gr-qc↗

arXiv2026

LeapAlign: Post-Training Flow Matching Models at Any Generation Step by Building Two-Step Trajectories

This paper focuses on the alignment of flow matching models with human preferences. A promising way is fine-tuning by directly backpropagating reward gradients through the differentiable generation process of flow matching. However, backpropagating through long trajectories results in prohibitive memory costs and gradient explosion. Therefore, direct-gradient methods struggle to update early generation steps, which are crucial for determining the global structure of the final image. To address this issue, we introduce LeapAlign, a fine-tuning method that reduces computational cost and enables direct gradient propagation from reward to early generation steps. Specifically, we shorten the long trajectory into only two steps by designing two consecutive leaps, each skipping multiple ODE sampling steps and predicting future latents in a single step. By randomizing the start and end timesteps of the leaps, LeapAlign leads to efficient and stable model updates at any generation step. To better use such shortened trajectories, we assign higher training weights to those that are more consistent with the long generation path. To further enhance gradient stability, we reduce the weights of gradient terms with large magnitude, instead of completely removing them as done in previous works. When fine-tuning the Flux model, LeapAlign consistently outperforms state-of-the-art GRPO-based and direct-gradient methods across various metrics, achieving superior image quality and image-text alignment.

cs.CV↗

arXiv2026

QPEs from Warped Disk Collisions with EMRIs: Brightness-Recurrence Diagram and Gravitational-Wave Follow-up

Quasi-Periodic Eruptions (QPEs) display correlated long/short and strong/weak patterns that remain unexplained by existing flat-disk collision models. We propose that these features arise from an extreme-mass-ratio inspiral (EMRI) colliding with a warped accretion disk, likely formed after a tidal disruption event. The warp modulates both recurrence time and burst energy, encoding the disk geometry -- and thus the spin of the central supermassive black hole (SMBH) -- into the X-ray light curve. We introduce the Brightness-Recurrence Diagram (BRD) to visualize this correlation, where QPE bursts trace an elliptical trajectory driven by the EMRI's apsidal precession; the tilt of this ellipse encodes whether the EMRI is prograde or retrograde relative to the SMBH spin. Applying this model to the prototypical QPE source GSN 069 successfully reproduces the observed patterns. The data are consistent with either a prograde stellar secondary or a retrograde stellar-mass black hole. In the stellar-mass black hole scenario, ongoing orbital decay could render the EMRI detectable by LISA within a few decades, facilitating gravitational-wave follow-up and independent multimessenger constraints on the system.

astro-ph.HE↗

arXiv2026

GenFirst: Generation Before Reconstruction for Stable End-to-End Latent Generative Modeling

Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on the frozen latent space. Since reconstruction-optimized latents are not necessarily generation-friendly, jointly training both models is an appealing alternative. However, direct end-to-end training remains challenging, as it is prone to latent collapse and faces a generation-reconstruction conflict. We revisit this problem by analyzing how different objectives shape the latent space and identify two key insights. First, the entropy term in the Kullback-Leibler divergence objective is essential for preventing collapse: reconstruction and prior fitting tend to shrink the posterior, while entropy preserves non-degenerate latent uncertainty. Second, reconstruction and generation exhibit asymmetric learning dynamics: reconstruction is fast and strongly supervised, whereas generation is slower and harder to optimize. Based on these insights, we achieve the first direct end-to-end training without latent collapse and propose GenFirst, a simple generation-before-reconstruction strategy. The generative objective first shapes the latent space under weak reconstruction pressure, after which reconstruction is progressively strengthened to recover visual details. We validate GenFirst with continuous autoregressive priors with exact likelihoods and SiT priors with implicit likelihoods. With our end-to-end objective and GenFirst, SiT achieves a gFID of 0.97 with CFG and 1.45 without CFG on ImageNet-256, while MMDiT reaches a GenEval score of 0.90 on text-to-image generation. Beyond image generation, we extend the framework to shared visual latents for generation and representation learning, and to continuous unified text-image generation. These results demonstrate the generality of stable end-to-end latent learning across generative priors and modalities.

cs.CV↗

arXiv2026

When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models

Data-poisoning backdoors pose a practical threat to the fine-tuning of large language models (LLMs). Most existing attacks bind an attacker-selected behavior to fixed tokens, phrases, scenarios, or syntactic structures. These discrete triggers provide concrete handles for defenses based on local token anomalies, pattern matching, or trigger recovery. We found that, \emph{under semantics-preserving rewriting, emotionally styled inputs form representation clusters distinct from their neutral counterparts}. Meanwhile, de-emotionalised controls move back towards the neutral distribution. This observation motivates our method \textbf{Paraesthesia}, a dynamic backdoor attack that encodes its triggering condition in an emotional style. Paraesthesia maps target emotions into a valence--arousal space, rewrites a small subset of clean samples, and retains semantically faithful rewrites for fine-tuning. Across instruction-following and classification tasks evaluated on four major LLMs, Paraesthesia achieves an attack success rate(ASR) above 98.25\%, while introducing only negligible degradation to clean utility across the vast majority of model-task setups. Surface feature controls and paired de-emotionalization experiments demonstrate that no examined token-level cue can fully account for the triggered behavior. ASR remains high after word-level filtering, sample clustering, and subsequent clean-update procedures, whereas a white-box decoding defense with access to a task-aligned clean reference provides a distinct mitigation path. These findings identify emotional style as a concrete backdoor trigger surface beyond fixed lexical and syntactic patterns.\par\smallskip \noindent \textcolor{red}{\textbf{WARNING: }\textnormal{This paper contains risk-related textual content.}}

cs.CL↗

arXiv2026

crossMoDA Challenge: Evolution of Cross-Modality Domain Adaptation Techniques for Vestibular Schwannoma and Cochlea Segmentation from 2021 to 2023

The cross-Modality Domain Adaptation (crossMoDA) challenge series, initiated in 2021 in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), focuses on unsupervised cross-modality segmentation, learning from contrast-enhanced T1 (ceT1) and transferring to T2 MRI. The task is an extreme example of domain shift chosen to serve as a meaningful and illustrative benchmark. From a clinical application perspective, it aims to automate Vestibular Schwannoma (VS) and cochlea segmentation on T2 scans for more cost-effective VS management. Over time, the challenge objectives have evolved to enhance its clinical relevance. The challenge evolved from using single-institutional data and basic segmentation in 2021 to incorporating multi-institutional data and Koos grading in 2022, and by 2023, it included heterogeneous routine data and sub-segmentation of intra- and extra-meatal tumour components. In this work, we report the findings of the 2022 and 2023 editions and perform a retrospective analysis of the challenge progression over the years. The observations from the successive challenge contributions indicate that the number of outliers decreases with an expanding dataset. This is notable since the diversity of scanning protocols of the datasets concurrently increased. The winning approach of the 2023 edition reduced the number of outliers on the 2021 and 2022 testing data, demonstrating how increased data heterogeneity can enhance segmentation performance even on homogeneous data. However, the cochlea Dice score declined in 2023, likely due to the added complexity from tumour sub-annotations affecting overall segmentation performance. While progress is still needed for clinically acceptable VS segmentation, the plateauing performance suggests that a more challenging cross-modal task may better serve future benchmarking.

eess.IV↗

arXiv2026

Seismic Acoustic Impedance Inversion Framework Based on Conditional Latent Generative Diffusion Model

Seismic acoustic impedance plays a crucial role in lithological identification and subsurface structure interpretation. However, due to the inherently ill-posed nature of the inversion problem, directly estimating impedance from post-stack seismic data remains highly challenging. Recently, diffusion models have shown great potential in addressing such inverse problems due to their strong prior learning and generative capabilities. Nevertheless, most existing methods operate in the pixel domain and require multiple iterations, limiting their applicability to field data. To alleviate these limitations, we propose a novel seismic acoustic impedance inversion framework based on a conditional latent generative diffusion model, where the inversion process is made in latent space. To avoid introducing additional training overhead when embedding conditional inputs, we design a lightweight wavelet-based module into the framework to project seismic data and reuse an encoder trained on impedance to embed low-frequency impedance into the latent space. Furthermore, we propose a model-driven sampling strategy during the inversion process of this framework to enhance accuracy and reduce the number of required diffusion steps. Numerical experiments on a synthetic model demonstrate that the proposed method achieves high inversion accuracy and strong generalization capability within only a few diffusion steps. Moreover, application to field data reveals enhanced geological detail and higher consistency with well-log measurements, validating the effectiveness and practicality of the proposed approach.

cs.LG↗

arXiv2026

Decoupling Policy Extraction for Offline Reinforcement Learning

Offline RL methods commonly jointly train the actor and critic, where the critic is used to guide the actor toward higher-value actions. This coupled learning process is well motivated in online RL, where an improved actor collects new data that can further update the actor and the critic. However, training data remains fixed in offline RL, making actor-side policy improvement unable to generate new data to validate or correct the critic. Moreover, retaining this coupled paradigm leads to two related challenges. Firstly, actor updates can drift toward high-valued but potentially out-of-distribution (OOD) actions and amplify critic overestimation. Secondly, conservative value estimation or behavior-cloning regularization creates a difficult trade-off between suppressing OOD actions and selecting high-value actions within the data-supported region. Motivated by this observation, we revisit the conventional offline RL paradigm and propose decoupling policy improvement from actor training. Specifically, we train the actor solely to model the behavior distribution and perform policy improvement at inference time by reranking multiple actor-generated proposals with a separately learned critic. We refer to this paradigm as the decoupled policy extraction paradigm. Under such paradigm, the actor provides behavior-supported action candidates, while the critic performs value-based selection within this candidate set. Extensive experiments show that the decoupled policy extraction paradigm outperforms both behavior cloning and jointly learned offline RL methods, while remaining effective even with a naive Q-learning critic.

cs.LG↗

arXiv2026

EndoLIFT: Language-Disambiguated Latent-Conditioned Rectified Flow for Bidirectional Endoscopic Control

Routine gastrointestinal endoscopy is intrinsically bidirectional: the instrument is advanced to reach target anatomy and later withdrawn or retroflexed for inspection, while an external cue may require earlier reversal. When the requested phase changes before the visual scene does, nearly identical observations can require opposite axial actions. We identify and formalize this ambiguity in bidirectional endoscopic control as intent aliasing. We propose EndoLIFT (Endoscopic Language-Instruction Flow with Trajectory Latents), a vision-language-action policy that combines explicit language-based intent conditioning with a latent-conditioned rectified-flow action expert. The policy receives RGB, a language instruction, and the previous-action state; a 32-D variational trajectory latent stochastically conditions continuous action-chunk generation. Controlled same-observation instruction swaps establish that language selects the axial mode, independently of whether the trajectory latent is present. Relative to the matched model without latent conditioning, EndoLIFT improves navigation-direction accuracy by 11.1 percentage points and reduces wrong-direction advance by 83\%. An architecture-controlled 1-bit mode-flag reference exhibits weaker canonical-anchor switching, while EndoLIFT retains 82.8\% intent-following accuracy across 44 held-out linguistic variants. In closed-loop evaluation, EndoLIFT improves overall success by 30 percentage points over EndoLIFT w/o VTL on both the seen colon phantom and the unseen lung and stomach phantoms, and completes 10/10 ex-vivo porcine-trachea trials. These results separate language-based intent selection from the trajectory latent's contribution to directional correctness and robust retraction.

cs.RO↗

arXiv2026

ACTS-SQL: Agentic and Critic-Oriented Tree-Structured SQL Correctness with Large Language Models

Large Language Models (LLMs) have been increasingly adopted in Text-to-SQL systems, yet SQL errors remain a major obstacle in real-world Text-to-SQL inference pipelines. Existing SQL correction approaches either rely on large-scale, high-quality training data with substantial overhead, or adopt single-path agentic workflows that are brittle to early mistakes and prone to error propagation. To develop a practical SQL correctness system for industrial scenarios, we present a training-free framework that formulates SQL correction as a plan-guided, tree-structured debugging process. By maintaining multiple correction strategies and enabling backtracking, the framework mitigates error accumulation during iterative refinement. We further integrate execution-based verification and clause-level diagnostic tools to support strategy pruning and precise error localization. We evaluate the system on the BIRD-Critic benchmark and observe consistent accuracy gains over strong LLM backbones and representative agent-based baselines, achieving a 9.42% improvement over the previous state-of-the-art method. The framework is also deployed in the Torch Log Service (TLS) of Volcano Engine to support an online Text-to-TLS API. In production, it improves execution accuracy from 36.77% to 53.61% on real user queries with a representative strong LLM backbone (GPT-5). These results demonstrate the effectiveness and stability of our approach in real-world deployments.

cs.AI↗

arXiv2026

Impact of higher-order modes on eccentricity measurement in binary black hole gravitational waves

We investigate the systematic biases in measuring orbital eccentricity for binary black hole (BBH) mergers that arise when higher-order modes (HOMs) of gravitational waves are neglected in waveform modeling. Using Bayesian inference with the state-of-the-art eccentric, spin-aligned, higher-mode effective-one-body model SEOBNRv5EHM, we reanalyze six previously suggested eccentric gravitational-wave events--GW190521, GW190620, GW190701, GW191109, GW200129, and GW200208\_222617. Comparing results with its dominant-mode-only counterpart SEOBNRv5E, we find no statistically significant HOM-induced bias in eccentricity for any of these events, including GW190521, whose eccentricity has been debated in the literature. To explore representative parameter regimes vulnerable to HOM omission, we perform a broad zero-noise injection campaign varying detector-frame total mass, mass ratio, eccentricity, inclination, and network SNR. We find that significant systematic biases ($Δ_e/σ> 1$) arise predominantly in systems with relatively high total mass ($M^{\rm det}\gtrsim120M_\odot$), highly asymmetric mass ratios ($q \gtrsim 4$), large inclinations ($θ_\textrm{JN} \gtrsim 30^\circ$), and high SNRs ($ρ^N_\textrm{mf}\approx50$). Notably, for quasicircular BBHs with approximately $M^{\rm det}\gtrsim140M_\odot$, our results suggest that neglecting HOMs may lead to strong false-positive evidence for nonzero eccentricity. By contrast, for lower-mass systems ($M^{\rm det}\sim100 M_\odot$), HOM exclusion produces negligible eccentricity biases. Our results demonstrate that although current eccentric candidates are not impacted by HOM omission, future eccentricity measurements--particularly for relatively massive, asymmetric, or edge-on systems--require HOM-inclusive waveforms to avoid substantial systematic errors.

gr-qc↗

arXiv2026

Space as an Interventional Invariant: Cross-Modal Predictive Geometry for Stratified Cities and Em-Spaced Intelligence

Space is a foundational concept across mathematics, physics, spatial cognition, urban science, and embodied intelligence, yet these fields often treat spatial structure either as a shared geometric container or as a collection of disconnected representations. Such approaches struggle to explain how heterogeneous sensory and urban processes can jointly reveal a common spatial structure, particularly when different modalities do not share the same metric or representation. This paper addresses this gap by defining space as an interventional invariant: the minimal relational structure that preserves local compatibility and the conditional laws of future observations under admissible actions. We develop a cross-modal predictive geometry that integrates local state spaces, modality-specific observation maps, an action groupoid, and a canonical predictive-state quotient, with explicit causal conditions for identifying interventional rather than merely observational structure. The key theoretical result shows that, under joint point separation, equivariance, and interventional faithfulness, the latent space is identifiable up to the centraliser of the intervention group, thereby reducing representational ambiguity to residual coordinate freedom. The framework is further extended to stratified urban systems using sheaf-valued representations, allowing geometric, physical, mobility, social, and economic layers to coexist without being reduced to a single metric. Synthetic experiments under noise evaluate equivariance, predictive sufficiency, holonomy, restriction-map recovery, cross-scale consistency, and context saturation. The resulting framework provides a unified and falsifiable foundation for spatial cognition, urban science, embodied AI, and em-spaced intelligence.

cs.LG↗

arXiv2026

Global Optimization and Inference-Time Region Grafting for Agentic Workflows

Recent advances in agentic workflow optimization automate workflow design through task-specific workflow search or input-conditioned architecture selection. However, they determine the workflow before execution and cannot adapt failed workflow regions using execution-time label-free quality signals. Naively enabling such inference-time adaptation through whole-workflow re-optimization would be computationally prohibitive. To tackle this challenge, we introduce GRAFT, which preserves a globally optimized workflow while locally replacing only selected regions for each input. Without parameter training, GRAFT evaluates region-level alternatives using label-free execution-quality signals and accepts only replacements that improve local quality while preserving workflow-level consistency, thereby enabling instance-wise adaptation without whole-workflow re-optimization. GRAFT applies without modification across a range of tasks spanning mathematical reasoning, code generation, and multi-hop and knowledge-intensive question answering. Under matched optimizer and executor settings, it improves over the strongest prior workflow-optimization method, MaAS, by 3.85 points on average. Replacing only the executor with a stronger model yields further gains without re-optimizing the global workflow. This suggests that an optimized workflow is not merely a static optimization artifact, but an adaptable execution policy that can evolve with inference-time feedback and stronger executors.

cs.CL↗

arXiv2026

Precision Joint Constraints on Cosmology and Gravity Using Strongly Lensed Gravitational Wave Populations

We present a Bayesian framework to jointly constrain the Hubble constant ($H_0$) and the post-Newtonian parameter ($γ$), a key indicator of deviations from general relativity, using the population characteristics of strongly lensed gravitational wave (GW) events from binary black hole mergers. Our method extracts cosmological and gravitational information directly from the statistical properties of lensed GW populations, without relying on electromagnetic counterparts of the GW events, waveform modeling, and resolved stellar kinematics of the lens galaxy. This establishes lensed GW statistics as a clean and independent probe of cosmic expansion and gravitational physics. Assuming a flat $Λ$CDM cosmology and simulating a GW population observed by the third-generation detector Einstein Telescope, we demonstrate that this method can achieve precision levels of $0.60\%\sim0.99\%$ for $H_0$ and $0.53\%\sim3.3\%$ for $γ$ with various priors of matter density, significantly outperforming existing joint constraints, which typically achieve $2\%$ precision on $H_0$ and $20\%$ precision on $γ$. These results highlight the potential of lensed GW population statistics as a robust and efficient tool for probing both the expansion history of the Universe and the nature of gravity.

gr-qc↗

arXiv2026

Co-Evolving LLM Evaluators and Policies via DynamicRubric

Post-training with evaluator feedback on policy-induced samples serves as a major mechanism for improving large language models. As policies improve, these sampled responses become close in quality. These close candidates create a bottleneck for policy optimization: collapsed relative evaluator score gaps yield weak or misleading policy supervision. We theoretically characterize why these gaps matter through a probability allocation view, showing that the directional gain of shifting probability mass from one response to another is exactly the evaluator score gap between them. This identifies relative score gaps as the policy optimization signals that guide updates. Motivated by this view, we propose DynamicRubric, a response-set-conditioned evaluator--policy co-evolution framework that generates weighted binary rubric items for each candidate set and aggregates the resulting judgments into response-level scores. In our experiments with 8B backbones, DynamicRubric improves evaluator performance and provides stronger policy supervision than baselines using a 70B reward model or a 235B static rubric generator. DynamicRubric-optimized policies also show gains on verifiable reasoning and coding tasks. A DynamicRubric-optimized model is fully deployed in WeChat Search's AI answering scenario, where it serves all online traffic across tens of millions of requests per day and improves key online metrics. These results suggest a principle for evaluator-guided post-training: evaluators should evolve with the policies they supervise.

cs.LG↗

arXiv2026

Breaking the Dark Sector Degeneracy with Nonparametric Expansion--Growth Reconstruction

The dark energy equation of state (EoS) and a possible dark-sector interaction are degenerate at the level of background expansion: the same expansion history may be interpreted as time-varying dark energy, energy exchange with dark matter, or a mixture of both. We introduce a data-driven, nonparametric expansion--growth framework that breaks this degeneracy by simultaneously reconstructing the dark energy EoS and the dark-sector coupling. Allowing the interacting matter density to evolve freely, we incorporate the coupling into the linear matter perturbation equation and reconstruct the expansion and growth histories using Gaussian processes, with hyperparameters marginalized in a Bayesian treatment. Applying this method to Pantheon+, cosmic chronometers, BAO measurements including DESI DR2, and redshift-space-distortion data, we infer both $w_{\rm de}(z)$ and the interaction history over $0\lesssim z\lesssim 2$, without assuming either a parametric EoS or a prescribed interaction form. We find no statistically significant evidence for either a nonzero interaction or dark energy dynamics: the reconstructed coupling and dark energy EoS remain consistent with the $Λ$CDM limit. Our results establish an expansion--growth consistency test for coupled dark-sector physics and provide a model-independent route to distinguish genuine dark energy dynamics from effective dark-sector energy exchange.

astro-ph.CO↗