arXiv subjects

Zhangyang Wang

Publications and source records attributed to Zhangyang Wang.

At least 19 recordsLinked to original sources

arXiv2026

Representation-Aligned Auxiliary Supervision for Language Model Adaptation

Language models exhibit strong reasoning capabilities, yet adapting them to structured domains remains challenging and can yield inconsistent outcomes. We identify representation compatibility, the extent to which a model effectively processes a representation for a structured task, as a key factor in adaptation. We study this in chess, which provides a controlled testbed with precise semantics, computable optimal actions, and multiple state representations, including a symbolic encoding (FEN) and a spatial format (ASCII). We find that models often process semantically equivalent inputs substantially differently, affecting both learning and generalization. Building on this observation, we propose representation-aligned auxiliary supervision, which uses environment-derived tasks expressed in compatible representations to improve adaptation to structured domains. Across models and representations, auxiliary supervision consistently improves optimal-move prediction relative to target-only training under identical target data. Tasks that expose environment dynamics provide larger and most consistent gains than surface-level or static supervision, while remaining competitive with substantially increasing the amount of target-task data. Moreover, ASCII-trained models transfer more effectively to FEN than FEN-trained models do to ASCII, even surpassing the FEN target-only baseline on FEN evaluation. The gains also extend beyond optimal-move prediction to open-ended, factually grounded commentary generation. Overall, our results show that auxiliary supervision in model-compatible representations can enable effective adaptation in structured domains.

cs.CL↗

arXiv2026

FinVerse: Toward More Realistic Evaluation for Financial Time-Series Forecasting

Time-series foundation models (TSFMs) are increasingly evaluated on broad benchmarks that apply uniform conventional forecast evaluation metrics across heterogeneous time-series. However, in the financial domain, decision-relevant evaluation requires going beyond conventional metrics to assess forecasts in terms of directional accuracy, cross-sectional ranking, and portfolio outcomes. To this end, we introduce FinVerse, a financial time-series benchmark, collecting a broad collection of 116,897 series with 171.1M observations. FinVerse organizes financial data according to their economic semantics and evaluates target series across point-wise forecasting, cross-sectional ranking, and portfolio backtesting. Across 43 public TSFMs, model rankings on FinVerse show only moderate agreement with those on the existing benchmark, GIFT-Eval (r=0.43), suggesting that generic forecasting benchmarks do not fully capture the aspects of forecast quality that matter for financial decision-making. Furthermore, substantial variation in model performance across financial time-series categories and evaluation metrics provides a more informative basis for model selection, enabling practitioners to identify suitable models according to their target series and downstream objectives.

cs.LG↗

arXiv2026

LLMs Can Get "Brain Rot": A Pilot Study on Twitter/X

We propose and test the LLM Brain Rot Hypothesis: continual exposure to junk web text induces lasting cognitive decline in large language models (LLMs). To unveil junk effects, we designed a novel controlled experiment on real Twitter/X corpora, by constructing junk and reverse-controlled datasets via two orthogonal operationalizations: M1 (engagement degree) and M2 (semantic quality), with matched token scale and training operations across conditions. Compared to the control group, continual pre-training of 4 LLMs on the junk dataset causes non-trivial declines (Hedges' g>0.3) on reasoning, long-context understanding, safety, and inflating "dark traits" (e.g., psychopathy, narcissism). The gradual mixtures of junk and control datasets also yield dose-response cognition decay: for example, under M1, ARC-Challenge with Chain-of-Thought drops 72.1 -> 57.2 and RULER-CWE 83.7 -> 52.3 as junk ratio rises from 0% to 100%. Error forensics reveal several key insights. First, we identify thought-skipping as the primary lesion in reasoning: models increasingly truncate or skip chains. Second, partial but incomplete healing is observed: scaling instruction tuning and clean continual pre-training improve the declined cognition, yet cannot restore baseline capability, suggesting persistent representational drift rather than format mismatch. Finally, we discover that the popularity, a non-semantic metric, of a tweet is a better indicator of the Brain Rot effect than the length in M1. Together, the results provide significant, multi-perspective evidence that social effects of data could be a causal driver of LLM capability decay in continual pre-training, thereby motivating routine "cognitive health checks" for deployed and evolving LLMs.

cs.CL↗

arXiv2026

Disentangling Task-Relevant Latent Dynamics for Reward Modeling

Reward design remains a major bottleneck for reinforcement learning in robotic manipulation. Handcrafted dense rewards can compensate for the limited guidance provided by sparse rewards, but require substantial task-specific reward engineering. Learning rewards from videos provides a scalable, data-driven alternative to manual reward design. However, existing video-based rewards are often computed in entangled visual spaces where task-relevant state changes are confounded with extraneous visual variation. As a result, the learned reward may respond to visual distractors instead of actual task execution. To address this limitation, we propose DiTaR (Disentangling Task-Relevant Latent Dynamics for Reward Modeling), a framework for learning dense rewards from heterogeneous videos. DiTaR learns a continuous task-relevant representation from multi-frame video latents through an asymmetric two-stage procedure that suppresses task-irrelevant visual variation. In the task-relevant latent space, we train a history- and language-conditioned rectified-flow model to capture task dynamics and derive a directional dense reward from temporal changes in dynamics compatibility. Together, these components enable reward modeling directly over task-relevant dynamics rather than generic visual variation. DiTaR substantially improves policy learning on MetaWorld-v3 and ManiSkill over video-based reward baselines, and transfers to real-world offline RL through reward relabeling. Reward analyses further show that DiTaR provides more reliable rewards.

cs.RO↗

arXiv2026

When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis

Financial decision-makers face more information than they can directly inspect, making context compression necessary. Yet when large language models (LLMs) compress financial source material, they can alter the investment judgment supported by the original source. We frame this problem as information fidelity: compression loses fidelity when it changes the decision induced by the source. In agentic systems, such losses may recur across intermediate steps and amplify throughout the decision process. Across financial filings and earnings-call transcripts, we find that LLM-based compression can produce fluent and factually plausible compressed contexts that nevertheless alter downstream decisions. We analyze two diagnostic patterns associated with fidelity loss: decontextualization, where salient evidence is retained but separated from the caveats and contextual qualifiers needed for correct interpretation, and model dependency, where different compressors expose different views of the same source. We then propose Agentic Context Compression, which generates multiple candidate compressions and audits their disagreements against the original source. Our results suggest that financial compression should be evaluated not only by efficiency or factuality, but also by its ability to preserve decision-relevant context.

cs.AI↗

arXiv2026

Data Efficient Any Transformer-to-Mamba Distillation via Attention Bridge

State-space models (SSMs) have emerged as promising alternatives to Transformers for sequence modeling. However, training competitive SSMs from scratch remains computationally intensive, and the ecosystem around them is far less mature than that of Transformers. Moreover, the architectural differences between SSMs and Transformers make it challenging to efficiently transfer knowledge from pretrained Transformers. In this work, we propose Cross-architecture distillation via Attention Bridge (CAB), a distillation framework that transfers attention-related representations from Transformer teachers to state-space student models. Unlike conventional knowledge distillation that supervises only final predictions, CAB enables token-level intermediate supervision through a lightweight bridge and flexible layer-wise alignment. By aligning Transformer attention-related representations with Mamba's token-dependent state projections, CAB facilitates efficient cross-architecture knowledge transfer without inference-time overhead. Experiments on image classification and language modeling demonstrate that CAB improves Transformer-to-SSM distillation, particularly under limited-supervision settings. Overall, CAB provides an efficient pathway for transferring Transformer representations to SSM-based models, helping bridge the gap between mature Transformer ecosystems and emerging SSM ecosystems.

cs.LG↗

arXiv2026

Restore What Matters: Lessons from Joint Restoration and Recognition

Recognition pipelines typically adopt a restore-then-recognize workflow, yet decades of experience show that generating visually pleasing images seldom translates to improved recognition. We propose a Joint Restoration-for-Recognition (JR$^2$) paradigm: restore only what downstream tasks truly require, with task signals dictating where, how much, and whether restoration is necessary. JR$^2$ rests on three pillars: (i) Physics, employing optics-accurate turbulence simulation, extensible to blur and noise, to ground restoration in real image formation; (ii) Neuroscience, drawing on selective attention and neuroplasticity to direct model capacity toward identity-critical regions and frames while bypassing already-clean inputs; and (iii) Vision & Learning, coupling recognition loss end-to-end through restoration and alignment so that low-level edits maximize high-level identity stability. Evaluations on IARPA-BRIAR show consistent improvements (e.g., TAR@0.01% FAR +0.6; FNIR@1% FPIR -2.5), while a quality gate skips ~70% of clean frames, reducing cost. Ablations confirm physics priors enhance realism, joint training prevents catastrophic forgetting, and selective restoration suffices in many cases. We conclude that better-looking images are neither necessary nor sufficient; restoration modules must be task-driven, selective, and physically aware. Code, pretrained models, and recipes are provided for integration.

cs.CV↗

arXiv2026

Revisiting Spectral Representations in Generative Diffusion Models

Diffusion models have shown remarkable performance on diverse generation tasks. Recent work finds that imposing representation alignment on the hidden states of diffusion networks can both facilitate training convergence and enhance sampling quality, yet the mechanism driving this synergy remains insufficiently understood. In this paper, we investigate the connection between self-supervised spectral representation learning and diffusion generative models through a shared perspective on perturbation kernels. On the diffusion side, samples (e.g., images, videos) are produced by reversing a stochastic noise-injection process specified by Gaussian kernels; on the spectral representation side, spectral embeddings emerge from contrasting positive and negative relations induced by random perturbation kernels. Motivated by this, we propose a self-supervised spectral representation alignment method to facilitate diffusion model training. In addition, we clarify how joint spectral learning can benefit diffusion training from a geometric perspective. Furthermore, we find that the optimization of the spectral alignment objective is in an equivalent form of diffusion score distillation in the representation space. Building on these findings, we integrate a spectral regularizer into diffusion training objectives to improve the performance of diffusion models on multiple datasets. Experiments across images and 3D point clouds show consistent gains in generation quality. Code is released at https://github.com/yuehaowang/spectral-reg-diffusion.

cs.LG↗

arXiv2026

Fund2Persona: A Framework for Building and Refining Financial Advisor Personas from Fund Disclosure Data

Demand for personalized financial advice is growing, yet current LLM-based advisors often fail to provide consistent and specialized guidance. Simple persona prompts rarely specify how a financial advisor should reason and often drift toward generic recommendations. We propose Fund2Persona, a framework that builds financial-advisor personas from real-world fund disclosures and refines them through an actor-scorer-patcher loop. We test whether the resulting personas can predict held-out portfolio changes and produce commentary consistent with fund managers' own explanations. They outperform generic baselines on both tasks. We further study two downstream diagnostics: market-scenario generation, where persona retrieval broadens plausible investment views, and multi-turn investor-advisor conversations, where matched personas give more specific and useful advice than a generic advisor. These results suggest that real fund data can bring manager-specific investment expertise to LLM advisors rather than merely changing an LLM's surface style.

cs.CL↗

arXiv2026

The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflows

Recent systems for optimizing prompts, programs, and ML workflows typically rely on explicit outer-loop controllers such as evolutionary search, bandits, or textual-gradient methods. We ask a fundamentally different question: how much of this search policy can be internalized by a single tool-using agent? We present ReASearch, a unified framework for reasoning-driven optimization in which the agent autonomously decides what to evaluate, how to diagnose failures, which edits to make, and when to verify or restart. Rather than serving only as a proposal generator guided by hand-designed heuristics, the agent actively analyzes outcomes, allocates budget, and refines its strategy over long horizons through persistent memory. With a shared agent loop and domain-specific tools, ReASearch instantiates the exact same scaffold to optimize prompts, programs, and ML workflows. Across 14 diverse tasks, it is competitive with and mostly better than specialized optimization systems, achieving gains of 2% to 40% over strong domain-specific baselines, and in some cases discovering solutions that improve on prior human best-known results. Crucially, we observe that complex search behaviors, which are typically implemented by explicit controllers, emerge naturally from the agent's reasoning process.

cs.AI↗

arXiv2026

Poisoning Agentic Alpha: Adversarial Vulnerabilities Across Roles and Architectures in Multi-Agent Trading Systems

LLM-based multi-agent trading systems, in which specialized agents collaborate through structured communication to produce trading decisions, are moving rapidly from research prototypes to live deployments that control real assets. The same inter-agent communication that makes them effective also exposes them: a corrupted signal can propagate to the final decision and translate into realized financial loss. Unlike prior attacks that presume privileged access to system internals, we restrict the adversary to what is practically reachable---the source data and prompts agents consume---yielding a low-barrier, and thus democratized threat model instantiated as role-specific adversaries. We present the first systematic empirical study in the financial domain to characterize how an adversarial signal enters a multi-agent trading system and how far it survives toward the decision. Along the role axis, we decompose a widely-used trading pipeline into four functional roles---Analyst, Researcher, Trader, and Risk Manager---and pair each with an attack matched to its interface. Along the structural axis, we evaluate four communication topologies under data- and agent-level attacks, using the Adversarial Signal Preservation Score (APS) as a post-hoc lens on why some designs are more robust than others. We conduct experiments across five assets, two backbones, and two target directions. A central finding is that no architecture is inherently robust. These findings provide insights for the future design of safer and more robust agentic trading systems.

cs.AI↗

arXiv2026

Correct Answers from Sound Reasoning: Verifiable Process Supervision for Language Models

Training language models to produce both correct answers and sound reasoning remains an open challenge. Reinforcement learning with verifiable rewards typically optimizes only final outcomes, which can improve task accuracy at the expense of reasoning quality, producing inaccurate, incomplete, or inconsistent traces. We propose verifiable process supervision (VPS), a post-training framework that jointly optimizes prediction accuracy and reasoning quality by supervising structured intermediate claims. We first apply supervised fine-tuning to induce a structured reasoning format, enabling deterministic extraction and verification of intermediate claims for process-level rewards. To address the heterogeneous difficulty of reasoning subtasks, we introduce adaptive weighting that prioritizes components with the largest remaining errors, creating an implicit curriculum. We evaluate VPS on chess as a controlled testbed where reasoning steps can be deterministically verified against engine signals. While outcome-only RL improves move accuracy, it sharply degrades reasoning quality, increasing win-rate error by up to 112% and reducing internal consistency by up to 69%. In contrast, VPS preserves accuracy while significantly improving reasoning quality, reducing win-rate error by up to 30% and restoring consistency to near saturation. A reasoning-space analysis further shows that, without a structured prior, outcome-only RL converges to budget-dependent shortcuts rather than sound multi-step reasoning. Beyond chess, we observe the same phenomenon on math reasoning, where outcome-only RL improves accuracy while degrading step-level arithmetic and consistency, whereas VPS maintains both. These results show that VPS enables language models to reason both accurately and reliably in verifiable domains.

cs.CL↗

arXiv2026

Lean Refactor: Multi-Objective Controllable Proof Optimization via Agentic Strategy Search

We present Lean Refactor, a plug-and-play retrieval-augmented agentic framework for multi-objective, controllable, and version-robust refactoring of Lean proofs. LLM-generated proofs are notoriously correct-but-verbose and brittle across library versions, yet existing refactoring works overlook three practical challenges: 1) Lean refactoring is natively multi-objective (proof length, compilation cost, and version compatibility are often in tension); 2) Lean repositories have fragile compatibility, whereas LLM releases are unaware of Lean/Mathlib versions; 3) Training-based pipelines require repeated fine-tuning with each new LLM release, scaling neither with model churn nor with Lean's release cycle. Lean Refactor steers a frozen agentic LLM with retrievals from a curated database of multi-objective refactoring strategies, each densely annotated with metadata such as supported Lean/Mathlib versions and expected compilation-cost reduction. Experiments show over $70\%$ token-level compression on competition benchmarks, over $20\%$ on research repositories, and up to $60\%$ compilation-time reduction, outperforming prior work and Claude Code. Version-filtered retrieval further improves compression on the target Lean version, and refactored miniF2F proofs exhibit stronger zero-shot version transfer to future Lean releases than their unrefactored counterparts.

cs.LO↗

arXiv2026

GradientStabilizer:Fix the Norm, Not the Gradient

Training instability in modern deep learning systems is frequently triggered by rare but extreme gradient-norm spikes, which can induce oversized parameter updates, corrupt optimizer state, and lead to slow recovery or divergence. Widely used safeguards such as gradient clipping mitigate these failures but require threshold tuning and indiscriminately truncate large updates. We propose GradientStabilizer, a lightweight, drop-in gradient transform that preserves the instantaneous gradient direction while replacing the update magnitude with a statistically stabilized estimate derived from running gradient-norm statistics. We prove that the resulting stabilized magnitude is uniformly bounded on spike steps, independent of the spike size, and show how this boundedness controls optimizer state evolution in adaptive methods. Across LLM pre-training (FP16), quantization-aware pre-training (FP4), ImageNet classification, reinforcement learning, and time-series forecasting, GradientStabilizer consistently improves training stability, widens stable learning-rate regions, and reduces divergence relative to clipping-based baselines, even substantially reducing Adam's sensitivity to weight-decay strength. Code will be released soon.

cs.LG↗

arXiv2026

When AI Writes, Who Gets Cited? Evidence of Citation Monoculture Across Language Models

As language models move from drafting prose to running literature-search agents with tool calls, fabricated references are becoming easier to catch and constrain. The harder failure begins after every candidate is real: different models may still select the same narrow subset, producing citation monoculture without any single citation being wrong. We isolate this effect on 120 real papers. Eleven models from three vendors choose at most ten papers from uniformly random panels of thirty, with real titles and abstracts but fabricated authors, reassigned years, and hidden venues and citation counts. Each run is compared with indifferent selection on the same panel and realized budget. All eleven models concentrate sharply: the top decile receives 23.3-30.2% of citations against 15.6% under the null, one component explains 68-73% of variation across their preference maps, and cross-vendor agreement nearly matches within-vendor agreement. Formalizing the task as fixed-budget subset selection, we turn these patterns into identifiable mechanisms: an exchangeability bound rejects a mapless selector for every model, a spectral decomposition explains why the best cross-fitted mixture still retains 55% of the excess, and a rarity theorem predicts the recursive competition effect we verify within panels. Controlled paraphrase, content-slot crossover, and design resampling attribute about 90% of GPT-5 mini's map variance to paper content. Eight domain experts selecting from the same blinded panels under the same cap show no comparable shared preference, while model concentration persists in selection-only mode. Even when every reference is real and every paper is equally visible, current language models impose a common content-level filter on scientific attention. Equalizing retrieval or mixing vendors is therefore insufficient; the shared preference map itself must be changed.

cs.DL↗

arXiv2026

Seeing Where to Deploy: Metric RGB-Based Traversability Analysis for Aerial-to-Ground Hidden Space Inspection

Inspection of confined infrastructure such as culverts often requires accessing hidden spaces whose entrances are reachable primarily from elevated viewpoints. Aerial-ground cooperation enables a UAV to deploy a compact UGV for interior exploration, but selecting a suitable deployment region from aerial observations requires metric terrain reasoning involving scale ambiguity, reconstruction uncertainty, and terrain semantics. We present a metric RGB-based geometric-semantic reconstruction and traversability analysis framework for aerial-to-ground hidden space inspection. A feed-forward multi-view RGB reconstruction backbone produces dense geometry, while temporally consistent semantic segmentation yields a 3D semantic map. To enable deployment-relevant measurements without requiring LiDAR-based dense mapping, we introduce an embodied motion prior that recovers metric scale by aligning predicted camera motion with onboard platform egomotion. From the metrically grounded reconstruction, we construct a confidence-aware geometric-semantic traversability map and evaluate candidate deployment zones under reachability constraints. Experiments on a tethered UAV-UGV platform demonstrate reliable deployment-zone identification in hidden space scenarios.

cs.RO↗

arXiv2026

Can Induced Emotion Bias LLM Behaviors in Sequential Decision Making?

As Large Language Models (LLMs) are increasingly deployed as autonomous agents in high-stakes domains, understanding contextual factors that may modulate their decision-making becomes critical. While LLMs are trained to perceive and resonate with users' emotions, it remains unclear whether induced emotion can influence their sequential decision-making. We investigate this question using the Iowa Gambling Task (IGT), a classic psychological paradigm for studying decision-making under uncertainty, combined with an imagination-based emotion induction procedure. We first validate the feasibility of this paradigm by confirming that LLMs can sense strong, distinguishable emotions from context and that LLM agents can learn from sequential interactions in a human-like pace. With the validated setup, we find that, different from humans, induced emotion does not significantly bias the decision dynamics of LLM agents on average. However, the effects of anger are conditioned: inducing anger makes LLM agents less sensitive to penalties for bad decisions, and in early stages of the game, anger can lower exploration, locking decisions into a few choices early. These findings reveal the subtle yet distinct effects of induced emotion on LLM decision-making compared to human behavior, and provide a tool for future research on affective modulation of LLM agents.

cs.CL↗

arXiv2026

OpenLongTail: Generative Scaling of Long-Tail Driving Data

Scaling robust driving policies is fundamentally bottlenecked by the scarcity of edge cases in curated datasets. While the real world continuously captures these critical events, such long-tail events remain underutilized when collected from heterogeneous sources. Specifically, diverse but valuable in-the-wild long-tail videos lack the full view coverage required for training policy models, often missing multi-view poses or originating solely from monocular dash cameras. This modality gap prevents these ubiquitous observations from being converted into scalable training data for long-tail generalization. We introduce OpenLongTail, an open-source generative data engine for scaling autonomous driving policies under long-tail events. To transform heterogeneous data sources into view-aligned and temporally coherent multi-view assets that are useful for policy learning, we develop a pose-informed extrapolative view synthesis pipeline that generates the missing views. We further enhance cross-view consistency and the temporal alignment for the newly generated views by injecting Plücker ray geometry into the scalable generation engine. By synthesizing heterogeneous long-tail data, we observe a significant improvement in closed-loop driving robustness in handling long-tail events. By measuring the extrapolative view synthesis and pose metrics, we validate the effectiveness of OpenLongTail in visual fidelity, cross-view consistency, and ego-trajectory recovery.

cs.CV↗