Search arXiv⌕ Search

arXiv subjects

Wentao Zhang

Publications and source records attributed to Wentao Zhang.

At least 127 records · Page 7Linked to original sources
arXiv2026

DetectZoo: A Unified Toolkit for AI-Generated Content Detection Across Text, Audio, and Image Modalities

The growing popularity and capacity of generative models have eroded the distinction between human and machine-generated content, motivating a growing body of work on detection across text, images, and audio. Most available detectors are either commercial software or, if open-source, come with incompatible codebases with bespoke preprocessing, evaluation protocols, and evaluation metrics, which make their adoption, fair comparison, and reproduction quite difficult. To address this critical gap, we introduce DetectZoo, a first-of-its-kind, extensible toolkit designed to provide a unified interface for AI-generated content detection across text, audio, and image modalities. DetectZoo standardizes the complete empirical pipeline, from data ingestion and preprocessing to model assessment, offering researchers a cohesive framework to benchmark state-of-the-art detectors systematically. By integrating diverse public datasets and baseline detection algorithms under a single, unified API, our toolkit facilitates rigorous and reproducible evaluation. DetectZoo provides reference implementations of 61 detectors, native loaders for 22 benchmark datasets, and a standardized evaluation pipeline that reports multiple metrics through a common interface. Each detector is self-contained yet accessible through the same interface, automatically caches pretrained weights, and reproduces the original published results. DetectZoo lowers the barrier to entry for multi-modal AI forensics, enabling researchers to identify performance gaps across domains and accelerating the development of robust, generalizable detection techniques. The open-source repository and comprehensive documentation are publicly available at https://github.com/sadjadeb/DetectZoo, and the package can be installed via pip install detectzoo.

cs.MM↗
arXiv2026

ANDES: Agent Native Data Evolving Synthesis Tool for Autonomous Instruction Alignment

AI agents are increasingly being tasked with automating AI research itself, particularly the critical post-training phase that transforms base LLMs into aligned assistants. However, recent evaluations reveal that even frontier agents struggle to perform this task. While the success of post-training fundamentally relies on acquiring high-quality data, relying on agents to autonomously curate targeted training datasets from the open web introduces severe challenges. Executing the long-horizon tasks of searching, filtering, and balancing data within noisy web environments frequently overwhelms an agent's limited context, ultimately leading to degraded dataset quality and suboptimal downstream training performance. To bridge this gap, we introduce Andes (Agent Native Data Evolving Synthesis), a framework that reimagines data generation as a plug-and-play \emph{agent skill}. Rather than forcing agents to devise complex data-gathering strategies from scratch, \textsc{Andes} provides an intelligent abstraction layer. By leveraging a self-evolving World Tree routing mechanism and actionable diagnostic reports, it allows trainer agents to dynamically steer data synthesis through an interactive, closed-loop interface. We demonstrate that under strict compute constraints, equipping foundationally weaker agents with Andes improves automated alignment, securing state-of-the-art performance on PostTrainBench and robust cross-task generalization. Our project is available at https://github.com/zzy1127/ANDES.

cs.AI↗
arXiv2026

Seeing Before Agreeing: Aligning Multi-Agent Consensus with Visual Evidence

Vision-language models (VLMs) have achieved strong performance on visual question answering (VQA). To mitigate individual hallucinations and blind spots, aggregating diverse perspectives via multi-agent collaboration has emerged as a promising paradigm. While this approach has shown great success in textual QA, its potential in the multimodal domain remains under-explored. Existing multi-agent VQA methods predominantly adapt text-centric protocols, focusing on textual discussions while ignoring the alignment of visual information. In this work, we reveal a key insight: answer-level agreement is insufficient for reliable multi-agent VQA; \textit{aligned visual evidence} -- shared support from the image regions agents rely on -- is essential for trustworthy consensus. To leverage this insight, we propose EAGLE (\textbf{E}vidence-\textbf{A}ligned \textbf{G}rounded mu\textbf{L}ti-agent r\textbf{E}asoning), a training-free evidence-centered framework for coordinating multiple VLM agents. EAGLE explicitly exposes each agent's grounding regions as visual evidence, enables mutual verification over the evidence, and uses evidence consistency to guide final decision-making. Experiments on six VQA benchmarks show that EAGLE achieves best average performance across domains while remaining lightweight, interpretable, and practical for deployment.

cs.CV↗
arXiv2026

AgentOrchestra: Orchestrating Multi-Agent Intelligence with the Tool-Environment-Agent(TEA) Protocol

Recent advances in LLM-based agent systems have shown promise on complex, long-horizon tasks, but existing agent protocols (e.g., A2A and MCP) do not adequately support lifecycle-aware coordination across agents, tools, and environments. To address this limitation, we introduce the \textbf{Tool-Environment-Agent} (TEA) protocol, a unified abstraction that models these components as first-class, versioned resources with explicit lifecycles. TEA supports end-to-end context and version management, improving traceability and reproducibility, while also enabling continual self-evolution of agent-associated components\footnote{Unless otherwise specified, \emph{agent-associated components} include prompts, memory/tool/agent/environment code, and agent outputs (solutions).}. Building on TEA, we present \projectname, a hierarchical multi-agent framework in which a central planner coordinates specialized sub-agents and dynamically extends capabilities during execution. Experiments on four challenging benchmarks, spanning expert-level agent tasks and scientific/mathematical reasoning, show that AgentOrchestra consistently outperforms strong baselines; in particular, it achieves 89.04\% on the GAIA Test set, placing it among the leading methods to the best of our knowledge. These results highlight the value of explicit protocol design and hierarchical orchestration for building more robust and adaptive multi-agent systems.

cs.AI↗
arXiv2026

One-Shot Klein Cutting Planes for Lipschitz Geodesically Convex Optimization in Hyperbolic Space

Motivated by the COLT 2023 open problem of Criscitiello, Martínez-Rubio, and Boumal on deterministic first-order methods for Lipschitz geodesically convex optimization on Hadamard manifolds, we study hyperbolic space \[ \HH^d_{-\kappaC^2} =\{X\in\R^{d+1}:\ipL{X}{X}=-1,\ X_0>0\}, \qquad \ip{U}{V}_X=\kappaC^{-2}\ipL{U}{V}. \] For every geodesically convex $M$-Lipschitz function \[ f:\bar B_{\HH}(x_0,r)\to\R,\qquad s=\kappaC r, \] we give a one-shot Klein cutting-plane method returning a queried point $\hat x$ such that \[ f(\hat x)-\min_{\bar B_{\HH}(x_0,r)}f\le \eps Mr \] after at most \[ \left\lceil 2d(d+1)\log\!\left(\frac{16\sinh s\cosh s}{s\eps}\right) \right\rceil \] oracle calls. For $d\ge2$, each localization step costs $O(d^2)$ arithmetic operations; for $d=1$, an interval variant gives the same oracle bound. Hence \[ N=O\bigl(d^2(s+\log(e/\eps))\bigr) =O\bigl(d^2ζ_s\log(e/\eps)\bigr), \qquad ζ_s=s/\tanh s . \] Compared with the constant-curvature construction associated with the COLT problem, this replaces chained curvature--accuracy dependence by additive dependence. The proof does not rely on convexity of the Klein pullback, which is generally only quasiconvex. Instead, every Riemannian subgradient halfspace becomes an exact Euclidean central cut: for $θ=\kappaC\dist(X,Y)$, \[ \ip{g}{\log_XY}_X =\fracθ{\kappaC^2\sinhθ}\ipL{g}{Y}, \] and tangency at $X$ converts $\ipL{g}{Y}\le0$ into \[ \gbar^{\mathsf T}(u-c)\le0,\qquad u=Φ(Y),\ c=Φ(X). \] Thus one fixed Euclidean ellipsoid localizes the hyperbolic ball, and curvature enters only through \[ \log\!\left(\frac{\sinh s\cosh s}{s\eps}\right) =\log(1/\eps)+2s-\log(4s)+O(e^{-4s}). \] The general Hadamard-manifold problem remains open.

cs.DS↗
arXiv2026

The Curse of Helpfulness: Inverse Scaling Law in Robustness to Distractor Instructions via DistractionIF

Large Language Models (LLMs) are increasingly deployed in agentic and retrieval-augmented generation (RAG) systems, where they must execute user-specified tasks over externally provided reference text. In practice, such context is often unstructured and contaminated with benign but instruction-like semantic noise, such as editorial comments and system traces, which should be treated strictly as data. We introduce DistractionIF, a benchmark designed to evaluate robustness against such distractor instructions in reference text. Across a broad range of models, we observe a consistent inverse scaling phenomenon: larger models are often less robust, with performance dropping by up to 30 points as scale increases. Mechanistically, our perplexity analysis reveals that scaling erodes the probabilistic boundary between robust and distracted behaviors, making models increasingly prone to over-interpreting noise as instructions. To address this, we demonstrate that reinforcement learning, specifically Group Relative Policy Optimization (GRPO), can restore this boundary, improving robustness by up to 15.5% without compromising general instruction-following capability. Our findings highlight a critical instruction-following robustness gap in reference-grounded tasks and establish reinforcement learning as a promising path for enforcing strict data-instruction separation at scale.

cs.AI↗
arXiv2026

Source-Grounded Semantic Reinforcement Learning for Low-Resource Target-Language Generation

Low-resource target-language generation is often limited by scarce parallel data, while high-resource source-language monolingual data is abundant but difficult to use with standard supervised fine-tuning. We propose Source-Grounded Semantic Reinforcement Learning (SG-SRL), a resource-utilization framework that converts source-language monolingual data into cross-lingual semantic supervision for target-language generation. SG-SRL performs reference-free reinforcement learning (RL) on source-language data using a cross-lingual semantic reward model, instantiated by a cross-lingual reranker that scores the semantic relevance between the source input and the target-language generation. While this induces severe verbosity-based reward hacking, a lightweight recovery stage using a small parallel corpus restores fluency, conciseness, and task format while preserving the semantic gains. Experiments on Chinese-to-Thai generation show that SG-SRL improves semantic grounding and factual coverage over cold-start SFT. Additional analyses on long-form transfer and Tibetan embedding-based rewards clarify the generalization behavior of SG-SRL and show that an encoder-based semantic reward can substitute for an LLM-based reranker in a realistic low-resource language setting.

cs.CL↗
arXiv2026

AlphaForgeBench: Benchmarking End-to-End Trading Strategy Design with Large Language Models

The rapid advancement of Large Language Models (LLMs) has led to a surge of financial benchmarks, evolving from static knowledge evaluation toward interactive trading simulations. However, existing frameworks for evaluating real-time trading largely overlook a critical failure mode: the severe behavioral instability of LLMs in sequential decision-making under financial uncertainty. Through extensive experiments, we show that when deployed as trading agents, LLMs exhibit extreme run-to-run variance, generate inconsistent action sequences even under deterministic decoding, and frequently produce irrational action flipping across adjacent time steps. We attribute these behaviors to the stateless autoregressive nature of LLMs, which lack persistent memory of prior actions, together with their sensitivity to continuous-to-discrete action mappings in portfolio allocation tasks. These deficiencies fundamentally undermine the reliability and reproducibility of many existing online and offline trading benchmarks. To address these limitations, we propose AlphaForgeBench, a principled evaluation framework that redefines LLMs as quantitative researchers rather than stochastic trading agents. Instead of producing discrete trading actions, AlphaForgeBench requires models to generate executable alpha factors and compose factor-based trading strategies grounded in financial knowledge. This paradigm decouples reasoning from execution mechanics, enabling deterministic and reproducible evaluation while remaining aligned with real-world quantitative research workflows. Extensive experiments across multiple state-of-the-art LLMs demonstrate that AlphaForgeBench eliminates execution-induced instability and provides a rigorous benchmark for evaluating financial reasoning, strategy formulation, and alpha discovery. Webpage at https://finbrain-lab-hkustgz.github.io/AlphaForgeBench

q-fin.TR↗
arXiv2026

TRACER: Turn-level Regret Matching with Inner Reinforcement Credit for Cooperative Multi-LLM Reasoning

Large language models increasingly rely on either reinforcement learning or multi-agent prompting to improve reasoning, yet these two paradigms remain difficult to combine. Directly applying single-agent reinforcement learning to multi-turn multi-agent systems faces following dilemmas: i) Sparse rewards, role-level free-riding and excessive training overhead. ii) Agents only imitate to collaborate. iii) Fixed collaboration protocol falls into oscillating local optimum. We introduce TRACER, a turn-level reinforcement framework for cooperative multi-LLM reasoning. TRACER separates collaborative decision making into a controller-regret layer, where controllers learn whether the agents should speak or skip the current round through regret matching, and a generation-credit layer, which optimizes proposer and reviewer utterances with role-specific GSPO rewards. This design i) assigns credit at the level of both action modes and generated utterances, thus avoiding free-riding and sparse rewards. We only expand the choices made by the controllers, thus greatly reducing computational cost of training. Moreover, ii) agents acquire collaborative capability as they learn when to utter and what to speak. Finally, iii) by designing binary actions ingeniously, we extend classical game theory established for finite action spaces to deep learning, thus achieving mathematically rigorous convergence. We train all local RL-style methods on the GSM8K training split and evaluate on held-out GSM8K, MATH500, and GPQA-Diamond to measure in-domain accuracy, cross-benchmark generalization, inference cost, and correction-preservation behavior. The resulting framework provides a compact and reproducible testbed for studying learned collaboration policies beyond fixed debate, voting, or aggregation protocols. Code is available at https://github.com/Shark-Forest/TRACER.

cs.AI↗
arXiv2026

OpenWorldLib: A Unified Codebase and Definition of Advanced World Models

World models have garnered significant attention as a promising research direction in artificial intelligence, yet a clear and unified definition remains lacking. In this paper, we introduce OpenWorldLib, a comprehensive and standardized inference framework for Advanced World Models. Drawing on the evolution of world models, we propose a clear definition: a world model is a model or framework centered on perception, equipped with interaction and long-term memory capabilities, for understanding and predicting the complex world. We further systematically categorize the essential capabilities of world models. Based on this definition, OpenWorldLib integrates models across different tasks within a unified framework, enabling efficient reuse and collaborative inference. Finally, we present additional reflections and analyses on potential future directions for world model research. Code link: https://github.com/OpenDCAI/OpenWorldLib

cs.CV↗
arXiv2026

Ecosystem service demand relationship and trade-off patterns in urban parks across China

Understanding public demand for urban ecosystem services (ES) is crucial for effective green space management, yet the intricate relationships and potential trade-offs among these diverse demands remain poorly understood. Previous studies have yielded inconsistent findings, often limited by small samples or reliance on indirect proxies. Here, we provide the first national-scale, direct assessment of the relationship among demands for nine urban park ES using a survey dataset comprising 20,075 responses across China and a point-allotment experiment that directly quantifies the trade-off patterns among service demands. We found particularly strong preferences among urban residents in China for air purification and recreation services, at the expense of other services. These preferences were further reflected in three distinct demand bundles: air purification-dominated, recreation-dominated, and balanced demands, each delineating a typical group of people with distinct representative characteristics. Socio-economic and environmental factors, such as age, environmental interest, and mean annual precipitation, significantly influence the trade-off intensity among service demands. Our study pioneers the direct, quantitative analysis of relationships among ecosystem service demands, and the results underscore the need for tailored urban park designs that address diverse service demands to sustainably enhance the quality of city life in China and beyond.

econ.GN↗
arXiv2026

Beyond Benchmark Islands: Toward Representative Trustworthiness Evaluation for Agentic AI

Agentic AI systems increasingly act through tool-augmented, multi-step workflows whose failures (unsafe tool use, unauthorised actions, social harm) carry deployment-level consequences. Evaluation practice remains fragmented across isolated benchmark slices, and "trustworthiness" is frequently invoked but rarely defined operationally. We argue the central limitation is twofold: (i) the absence of a measurable specification of what agent trustworthiness means, and (ii) the lack of a principled notion of representativeness allowing assessment over a socio-technical scenario distribution rather than disconnected benchmark instances. We address (i) by defining agentic trustworthiness as a five-property profile (Reliability, Robustness, Safety, Social-Ethical Alignment, Operational Integrity) grounded in current AI risk frameworks, and (ii) with the Holographic Agent Assessment Framework (HAAF), which measures this profile over a scenario manifold through static policy analysis, sandbox simulation, social-ethical alignment assessment, and distribution-aware sampling, connected through an iterative Trustworthy Optimization Factory that converts red-team diagnoses into blue-team interventions. Our contributions are: (1) an operational five-property definition of agentic trustworthiness; (2) a distribution-aware scenario-sampling framework that surfaces property-level trade-offs invisible to scalar leaderboards; and (3) a cross-family transfer experiment in which interventions designed from a single focal model generalise -- without per-model or per-scenario tuning -- to 13 systems from seven model families (Llama, Mistral, Kimi, GLM, Qwen, GPT, DeepSeek) on a 100-scenario suite, where all 13 systems improve and two reach a perfect risk-weighted profile, establishing HAAF's Factory as a model-agnostic deployment-readiness pipeline. Code: https://github.com/TonyQJH/haaf-pilot

cs.CL↗
arXiv2026

LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent Reasoning

Joint audio-visual reasoning is essential for omnimodal understanding, yet current multimodal large language models (MLLMs) still struggle when reasoning requires fine-grained evidence from both modalities. A central limitation is that explicit text-based chain-of-thought (CoT) compresses continuous audio-visual signals into discrete tokens, weakening temporal grounding and shifting intermediate reasoning toward language priors. We argue that a unified latent space is a better medium for such reasoning because it preserves dense sensory information while remaining compatible with autoregressive generation. Based on this insight, we propose \textbf{LatentOmni}, a cross-modal reasoning framework that interleaves textual reasoning with audio-visual latent states. LatentOmni introduces feature-level supervision to align latent reasoning states with task-relevant sensory features and uses Omni-Sync Position Embedding (OSPE) to maintain temporal consistency between latent audio and visual states. We further construct \textbf{LatentOmni-Instruct-35K}, a dataset of audio-visual interleaved reasoning trajectories for supervising latent-space reasoning. Comprehensive evaluation across multiple audio-visual reasoning benchmarks demonstrates that LatentOmni achieves the best performance among the evaluated open-source models and consistently outperforms the Explicit Text CoT baseline, supporting latent-space joint reasoning as a promising path toward stronger omnimodal understanding.

cs.CL↗
arXiv2026

VGGT-Edit: Feed-forward Native 3D Scene Editing with Residual Field Prediction

High-quality 3D scene reconstruction has recently advanced toward generalizable feed-forward architectures, enabling the generation of complex environments in a single forward pass. However, despite their strong performance in static scene perception, these models remain limited in responding to dynamic human instructions, which restricts their use in interactive applications. Existing editing methods typically rely on a 2D-lifting strategy, where individual views are edited independently and then lifted back into 3D space. This indirect pipeline often leads to blurry textures and inconsistent geometry, as 2D editors lack the spatial awareness required to preserve structure across viewpoints. To address these limitations, we propose VGGT-Edit, a feed-forward framework for text-conditioned native 3D scene editing. VGGT-Edit introduces depth-synchronized text injection to align semantic guidance with the backbone's spatial poses, ensuring stable instruction grounding. This semantic signal is then processed by a residual transformation head, which directly predicts 3D geometric displacements to deform the scene while preserving background stability. To ensure high-fidelity results, we supervise the framework with a multi-term objective function that enforces geometric accuracy and cross-view consistency. We also construct the DeltaScene Dataset, a large-scale dataset generated through an automated pipeline with 3D agreement filtering to ensure ground-truth quality. Experiments show that VGGT-Edit substantially outperforms 2D-lifting baselines, producing sharper object details, stronger multi-view consistency, and near-instant inference speed. The project page is https://chriszkxxx.github.io/VGGT-Edit/.

cs.CV↗
arXiv2026

DataPrep-Bench: Benchmarking LLMs as Training Data Preparators

The quality of training data fundamentally determines the capabilities of large language models (LLMs), yet no unified benchmark exists to measure how well LLMs, agents, and data-centric workflows actually prepare training data end to end. We view LLM-driven data preparation as comprising two complementary capabilities: data construction, which transforms raw sources into supervised training data, and data quality evaluation, which predicts the training value of candidate datasets before downstream training; throughout, "quality" refers to downstream training utility rather than surface-level textual properties. We introduce DataPrep-Bench, the first unified benchmark that jointly evaluates both capabilities under a shared downstream-grounded protocol over six domains and multiple base models. For data construction, methods consume identical raw sources and are scored by fine-tuning a base model on their outputs jointly with Dolly-15k; alongside this track we release Data-Construction-Skill, a skill-guided agent that lifts the Dolly-only baseline by nearly 20 points absolute on Llama-3.1-8B Finance and is competitive with the strongest agent- and DataFlow-based methods in knowledge-extraction-dense domains. For data quality evaluation, scoring functions are scored by Pearson correlation with downstream performance on a shared candidate pool; we release the Distributional Alignment Score (DAS), a distribution-based evaluator that uses MMD between a candidate dataset and a domain proxy. DAS attains the strongest cross-model correlation in four of six domains and is the only metric clearing r > 0.70 simultaneously in Math, Science, and Medical, outperforming existing quality-, diversity-, and heuristic-based evaluators. DataPrep-Bench provides a unified, downstream-grounded framework for measuring progress on both capabilities as co-equal targets of LLM-driven data preparation.

cs.LG↗
arXiv2026

Reinforcement Learning with Semantic Rewards Enables Low-Resource Language Expansion without Alignment Tax

Extending large language models (LLMs) to low-resource languages often incurs an "alignment tax": improvements in the target language come at the cost of catastrophic forgetting in general capabilities. We argue that this trade-off arises from the rigidity of supervised fine-tuning (SFT), which enforces token-level surface imitation on narrow and biased data distributions. To address this limitation, we propose a semantic-space alignment paradigm powered by Group Relative Policy Optimization (GRPO), where the model is optimized using embedding-level semantic rewards rather than likelihood maximization. This objective encourages meaning preservation through flexible realizations, enabling controlled updates that reduce destructive interference with pretrained knowledge. We evaluate our approach on Tibetan-Chinese machine translation and Tibetan headline generation. Experiments show that our method acquires low-resource capabilities while markedly mitigating alignment tax, preserving general competence more effectively than SFT. Despite producing less rigid surface overlap, semantic RL yields higher semantic quality and preference in open-ended generation, and few-shot transfer results indicate that it learns more transferable and robust representations under limited supervision. Overall, our study demonstrates that reinforcement learning with semantic rewards provides a safer and more reliable pathway for inclusive low-resource language expansion.

cs.CL↗
arXiv2026

Generalized Boundary FDR Control under Arbitrary Dependence: An Approach on Closure Principle

False discovery rate (FDR) is a cornerstone of modern multiple testing. However, it often fails to guarantee the reliability of "marginal" discoveries that lie at the boundary of the rejection set, which are often crucial in high-precision applications. While recent works (Soloff et al., 2024; Xiang et al., 2025) introduced the boundary false discovery rate (bFDR) to control the error probability at the marginal discovery, their method relies on restrictive assumptions such as independence or specific prior distributions. In this paper, we first propose $k$-bFDR, a novel generalization that controls the error probability of the $k$ least significant discoveries. We then provide a systematic investigation into the theoretical relationship between $k$-bFDR and existing error metrics. Furthermore, building upon the closure principle, we develop Domino, a unified framework that guarantees $k$-bFDR control under arbitrary dependence, applicable for both p-values and e-values. We prove the theoretical validity of the proposed Domino algorithm and demonstrate through extensive numerical experiments that it consistently achieves rigorous $k$-bFDR control while identifying trustworthy marginal discoveries. Analyses of real data reveal that $k$-bFDR control yields higher-quality rejection sets with greater practical significance.

stat.ME↗
arXiv2026

FLARE: Full-Modality Long-Video Audiovisual Retrieval Benchmark with User-Simulated Queries

As video becomes increasingly central to information dissemination and multimodal large language models (MLLMs) continue to advance, evaluating video retrieval has become increasingly important. In realistic search scenarios, this requires matching short user queries to long-form content using both visual and auditory evidence. Yet existing retrieval benchmarks are still dominated by short clips, single modalities, and caption-based evaluation. We introduce FLARE, a full-modality long-video audiovisual retrieval benchmark with user-simulated queries. Built from 399 carefully screened Video-MME videos (10--60 min, 225.4 h) to ensure source quality and diversity, FLARE contains 87,697 clips annotated with vision, audio, and unified audiovisual captions, together with 274,933 user-style queries. Cross-modal queries are further filtered by a hard bimodal constraint, requiring retrieval to fail under either modality alone but succeed when both are combined. FLARE evaluates models under two regimes, caption-based and query-based retrieval, across vision, audio, and unified audiovisual settings. Experiments with 15 representative retrievers show that user-style queries substantially change model behavior, strong caption-based performance does not always transfer to query-based retrieval, and audio--language alignment remains a key bottleneck for unified audiovisual retrieval. Our code and data are released at https://flarebench.github.io/

cs.MM↗