arXiv subjects

Chao Li

Publications and source records attributed to Chao Li.

At least 19 recordsLinked to original sources

arXiv2026

Reliability, validity, and diagnostic evidence for multi-model LLM short-answer scoring

Large language models (LLMs) are increasingly used or proposed for educational scoring, but single-model and single-run evaluations provide limited evidence for assessment use. Short-answer scoring requires evidence about reliability, validity, severity, diagnostic value, and failure cases. This study evaluated repeated multi-model OCG-PRES guided LLM scoring for short-answer assessment. The analysis used 996 SciEntsBank responses. GPT, DeepSeek, and Qianwen each scored every response across three independent runs using five OCG-PRES dimensions: concept coverage, relation accuracy, reasoning completeness, contradiction control, and domain relevance. Scores were evaluated against official binary and five-category labels and compared with non-LLM baselines based on answer length, Jaccard keyword overlap, TF-IDF cosine similarity, and a combined traditional logistic model. Repeated-run reliability was high for all models, with ICC(3,k) = .977 for GPT, .992 for DeepSeek, and .981 for Qianwen. DeepSeek was the most stable across runs. GPT showed the strongest official-label alignment by AUC (.909), while Qianwen was stricter, with higher precision but lower recall under the fixed threshold = 3.0 rule. OCG-PRES scores followed expected diagnostic patterns across five official categories and outperformed all non-LLM baselines in AUC and F1. Repeated multi-model OCG-PRES scoring provides reliability, validity, and diagnostic evidence for LLM-assisted short-answer scoring. The findings support cautious, evidence-based use as a scoring support tool rather than a replacement for human judgement.

cs.CL↗

arXiv2026

Exploring the Potential of Contrastive Language-Image Pre-training for Multi-Source Remote Sensing Data

Contrastive language-image learning (CLIP) has become a key paradigm for remote sensing vision-language understanding. However, existing remote sensing contrastive learning methods are mostly built on RGB-oriented CLIP architectures, making it difficult to exploit heterogeneous sensors such as SAR, multi-spectral imaging (MSI), and hyperspectral imaging (HSI). To address this limitation, we propose OmniRSCLIP, an end-to-end contrastive learning framework that supports multi-source sensor inputs for remote sensing vision-language modeling. The key idea is to extend CLIP beyond its fixed RGB input interface without breaking the pretrained visual knowledge. To this end, OmniRSCLIP introduces Spectral-Spatial Basis Decomposition (SSBD), which formulates arbitrary-channel adaptation as a basis recomposition problem: pretrained CLIP patch embeddings provide transferable spatial bases, while wavelength-conditioned coefficients span sensor-specific embedding kernels within a constrained visual prior space. This design avoids forcing heterogeneous sensors into a fixed-channel input space, while aligning them in a unified image-text semantic space. We further introduce a spectral-context-aware mask-based contrastive learning scheme to suppress modality-specific redundant features and enhance fine-grained image-text alignment. Finally, to support multi-modal training, we construct OmniRS5M, the first large-scale remote sensing image-text corpus covering RGB, SAR, MSI, and HSI. Experiments on retrieval, zero-shot classification, and semantic localization show that OmniRSCLIP preserves strong RGB-domain performance while effectively extending CLIP to heterogeneous remote sensing modalities.

cs.CV↗

arXiv2026

Effects of the Direct Space Charge on Transverse Coupled-Bunch Instabilities

The ultralow transverse emittance of diffraction-limited electron storage rings leads to high beam densities, making direct space-charge effects increasingly relevant despite their relativistic suppression at high energy. In this work, we investigate the impact of direct space charge on transverse coupled-bunch instabilities, using PETRA-IV as a representative example. A Vlasov solver, multi-bunch particle tracking, and analytical calculations in normalized phase space are combined to study the coupled-bunch mode spectrum, the nonlinear incoherent tune spread, and the associated Landau damping. Particle-tracking simulations show that the space-charge-induced tune spread can provide sufficient Landau damping to stabilize higher-order head-tail modes. Landau stability diagrams constructed from the analytical tune spread of a frozen bi-Gaussian beam predict the stabilization of these higher-order modes, in agreement with the tracking simulations. These results demonstrate the important role of direct space charge in transverse coupled-bunch dynamics. The analysis framework developed here can also be extended to collective-instability studies in both electron and hadron accelerators.

physics.acc-ph↗

arXiv2026

Lightweight Adaptation of General-Purpose VLMs for Multispectral and SAR Image Understanding

General-purpose vision-language models (VLMs) now support strong visual recognition, instruction following, and generation. However, most pretrained visual encoders are built around three-channel natural images and do not directly accommodate observations such as native multispectral measurements or synthetic aperture radar (SAR). Adapting VLMs to these sensors typically requires dedicated encoders and domain pretraining, slowing the reuse of stronger general-purpose checkpoints. We show that the multi-image interface of general-purpose VLMs offers a lightweight alternative. Our protocol renders each observation as five optical views and one SAR view, names them in the prompt, and adapts the language network and selected visual transformer blocks with LoRA. This exposes band composites, spectral indices, and radar backscatter through an existing visual interface. For land-cover recognition, structured supervision couples predicted classes with sensor evidence. We further construct preference pairs in which a true label is omitted while its supporting evidence is retained, encouraging complete predictions that remain consistent with the observations. On a balanced six-class land-cover benchmark derived from BigEarthNet-v2, the adapted Qwen3-VL reaches 0.8275 micro F1. The same input and adaptation protocol improves all four tested VLM architectures and transfers to Sen1Floods11 flood verification and BigEarthNet.txt captioning. Image removal and mismatch controls show that the adapted models use the supplied sensor observations. Together, these results demonstrate that VLMs can be repurposed for multispectral and SAR tasks through rendered inputs and compact LoRA adaptation, without training a new foundation model.

cs.CV↗

arXiv2026

EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents

Vision-language-action (VLA) models map visual observations and language instructions directly to robot actions, but long-horizon tasks require more than action prediction. An agent must coordinate perception, planning, execution, progress verification, and recovery as the physical state evolves. An action prediction or a model-generated skill decision does not, by itself, guarantee that the proposed operation is valid in the current state or that its outcome will be verified. We propose EmbodiedSkills, a unified framework that treats each skill decision as an execution proposal: the runtime checks its prerequisites before execution and verifies the outcome afterward. A shared executable-skill interface connects high-level skill selection, bounded low-level VLA execution, and post-action verification within a single agent loop. Because this interface remains fixed, low-level VLA policies can be replaced or adapted without changing the agent loop. The interface also records planning, execution, verification, and recovery events as structured trajectories, which provide supervision for individual components and can support optional online adaptation when interactive feedback is available. We instantiate EmbodiedSkills with Qwen3-VL and OpenPI/pi0.5 on RoboTwin 2.0 and LIBERO. Task-adapted low-level VLA policies achieve an average success rate of 86.20% across 50 RoboTwin 2.0 tasks and 97.40% across the four LIBERO suites. These results establish the execution performance of the task-adapted low-level VLA policies used in EmbodiedSkills. On four memory-dependent RMBench tasks, the same task-adapted execution approach achieves 12.5% average success. The framework provides a trainable and inspectable agent layer for turning these policies into closed-loop embodied systems.

cs.RO↗

arXiv2026

Why Organizational Rules Fail AI: O-I-B-A-R and the Externalization of Decision Boundaries

AI systems increasingly enter organizations through policies, procedures, playbooks, prompts, and other explicit representations of work. Yet formal descriptions often differ from situated practice, and captured know-what can omit the contextual know-how experts use when judgments are uncertain. We argue that a recurring class of organizational AI failures arises partly from a knowledge representation problem at the sociotechnical interface: the AI receives the procedure, while the organization operates on the procedure plus negative boundaries, runtime judgments, responsibility assignments, and learning history. We introduce O-I-B-A-R (OPEN, IS, BUT, ACTION, RESULT), a scaffold for externalizing these missing decision boundaries. IS records when a judgment holds. BUT records a concrete failure containing information beyond the logical negation of IS. Comparable success and failure cases are decomposed toward a minimally sufficient changing variable, which becomes a value-bearing decision dimension. A suspension represents the state in which the dimension is known but its current value is unresolved, specifying what must be measured, asked, retrieved, or escalated to a human. RESULT confirms a boundary, shifts a threshold, or exposes a new dimension. Incidents can generate new dimensions, unresolved values can define human-AI handoffs, and feedback can expand the decision space. We also identify a sociotechnical tension: durable and attributable failure histories can suppress the candor on which useful boundary knowledge depends. Externalization must therefore be designed as an organizational intervention with real costs and incentives.

cs.CY↗

arXiv2026

MemFuse: Multi-Source Memory Fusion from Fragmented Observations

Long-term memory is essential for agents that operate across extended interactions, yet existing memory systems and benchmarks predominantly focus on single-source textual histories. In realistic settings, however, relevant information is often fragmented across applications and devices, as well as across users and time, requiring agents to integrate dispersed observations into coherent episodic memories while preserving their source provenance. To address these gaps, we introduce **MemFuseBench**, a benchmark for *multi-source memory fusion*. MemFuseBench is built with a Scene-to-Sensor pipeline that synthesizes controllable scenarios into source-tagged observations, evidence-grounded questions, and adversarial distractors. It enables systematic evaluation of temporal reasoning, cross-source evidence fusion, and robustness to noise. We further propose **MemFuse**, a structured memory system that preserves source-level evidence in event-layer atomic memory and organizes related atomic events into cluster-layer fused memory within a causal fusion graph. During retrieval, MemFuse retrieves and organizes related evidence fragments while maintaining traceability to original source events. Experiments on MemFuseBench show that MemFuse achieves the best overall performance among the evaluated memory systems under all three LLM settings and consistently improves performance on questions requiring cross-source evidence fusion.

cs.CL↗

arXiv2026

D$^2$ACCI: A Dual-Loop Diagnostic Protocol for Evidence-Preserving Agent Memory

Memory is a key capability of LLM agents. Persistent memory extends this across sessions---enabling recall, revision, and personalization. Yet its multi-stage pipeline (ingestion, retrieval, filtering, generation) makes failures difficult to localize: end-to-end evaluation reveals that an error occurred, but not which stage caused it. Existing evaluations often report aggregate performance without paired statistical comparisons, slice-level non-regression checks, or stage-level diagnostic traces. We propose D$^2$ACCI (Diagnostic-Driven Artifact-based Closed-loop Controlled Iteration), a dual-loop protocol whose outer diagnostic gate promotes, feature-flags, or rejects memory interventions based on paired evidence, protected-slice monitoring, and trace-level localizability. We further introduce DCR, a graded observability metric that measures whether failures remain localizable, and D$^2$ACCI-Eval, a reusable artifact for gate replay. We instantiate the protocol in MemStack and evaluate on three public benchmarks, achieving 93.59% on LoCoMo, 90.93% on LongMemEval, and 57.20% on PersonaMem-V2. Five paired ablations show that supplement extraction, session-memory retrieval, and Forget Guard yield statistically significant gains (+1.9 to +3.7pp, all p $\le$ .003). In contrast, BM25/RRF is retained as a monitored feature flag---a distinction invisible to aggregate-only evaluation. A diagnostic audit shows enriched traces substantially improve root-cause agreement over result-only relabeling. Diagnostic artifacts reach 98--100% DCR@3 versus 0% for results-only logs. These results establish that robust memory-system iteration demands traceable, statistically grounded, and regression-aware evidence---exactly the gap D$^2$ACCI fills.

cs.AI↗

arXiv2026

Sharp Neumann eigenvalue estimates and $C^2$ elliptic regularity in non-obtuse polyhedral domains

For integers $n\ge 1$, consider assertions: $\mathbf{(P_n)}$: Let $\Omega \subset S^n$ be a spherical domain enclosed by totally geodesic $S^{n-1}$'s with non-obtuse dihedral angles. Then in $[0,2(n+1)]$, its Neumann spectrum can only take values among $\{0,n,2(n+1)\}$. Moreover, $n$ is a Neumann eigenvalue if and only if the corresponding eigenfunction is the restriction of a linear function in $\mathbb{R}^{n+1}$, while $2(n+2)$ is a Neumann eigenvalue if and only if the corresponding eigenfunction is restriction of a quadratic polynomial in $\mathbb{R}^{n+1}$. $\mathbf{(Q_n)}$: A weak solution $u$ to $\Delta u = f$ with the Neumann boundary condition, with $f$ H\"older continuous, in a conical polyhedral domain $\Omega$ in $\mathbb{R}^n$ with non-obtuse dihedral angles, is in $C^{2,\alpha}_{loc}(\overline{\Omega})$. We prove the implications \[\mathbf{(Q_n)} \Rightarrow \mathbf{(P_n)},\qquad \mathbf{(P_n)}\Rightarrow \mathbf{(Q_{n+1})}.\] Consequently, both assertions hold in all dimensions. These give the optimal Neumann eigenvalue lower bound and $C^2$ elliptic regularity in non-obtuse Riemannian polyhedral domains.

math.DG↗

arXiv2026

A continually expandable foundation model for brain MRI

Brain magnetic resonance imaging (MRI) is central to neuroscience and clinical assessment, but models are commonly developed for individual diseases, populations or imaging protocols. Foundation models promise more general representations, yet they are usually pretrained once and can lose earlier capabilities when updated with new data. Here we show that Alcmaeon, a three-dimensional brain MRI foundation model pretrained without manual labels on more than 425,000 volumes and derived imaging maps, can be expanded sequentially across clinical domains. Alcmaeon combines volumetric encoding and latent diffusion generation with Graph-Blueprint Pruning (GBP), which protects network modules important to earlier domains while leaving the remaining capacity trainable. Across expansion from healthy ageing and neurodegeneration to developmental, psychiatric and tumour imaging, GBP showed less forgetting than sequential adaptation and elastic weight consolidation across voxel-level reconstruction measures, with its largest advantage after adaptation to tumour imaging. The blueprints provided an inspectable record of how model capacity was protected and reused. Representations from different model levels supported image synthesis, disease classification, survival modelling and postoperative prediction, although no single representation was optimal for every task. These findings provide a route towards brain MRI foundation models that can grow with emerging data while retaining earlier capabilities.

cs.CV↗

arXiv2026

GRACE: Generative Recommender Acceleration Engine for Real-Time Ads Retrieval

Productionizing generative recommenders for high-volume, real-time ads retrieval creates two serving challenges: eligibility, ensuring that each generated ad is eligible for the request under the advertiser's audience targeting rules, and compute, which requires meeting strict latency and GPU cost requirements while remaining capable of generating thousands of ads per request with wide-beam decoding. This paper presents GRACE, a serving system for ads generative retrieval that addresses both challenges. For eligibility, GRACE introduces Generative Target Matching (GTM), which extends catalog-valid constrained decoding with personalized filtering over Semantic ID (SID) prefixes using bitmask and Bloom filter matchers derived from targeting rules. SID-level GTM improves final ad-level target matching pass rate from 23.55% to 40.42% over constrained decoding alone. For compute-cost and latency, GRACE targets encoder-decoder Transformers, which are more lightweight than LLMs. It redesigns the decoder around the wide-beam, short-sequence regime, covering attention kernels, KV cache, and beam search optimizations. On NVIDIA GH200, compared with the faster of FlashAttention-2 and FlashAttention-3 baselines, GRACE improves cross-attention latency by 68.0 times and self-attention latency by 23.4-25.8 times across decode steps. Together, these changes reduce decoder latency by 11.1 times, keeping ads generative retrieval within latency and compute requirements.

cs.IR↗

arXiv2026

Attend to Your Own Thoughts: Breaking the Barrier for Post-Training Quantization of Reasoning LLMs through the Lens of 1.58-Bit Quantization

We propose ScaleQ-1.58, a scalable ternary post-training quantization (PTQ) framework for reasoning LLMs. Its core insight stems from an empirical finding: although modern LLMs are typically trained to exhibit chain-of-thought reasoning capabilities, in the PTQ regime, even the latest CAT-Q method based on learning-based differentiable ternarization still leads to performance collapse on challenging mathematics and coding tasks when using conventional calibration schemes that ignore the model's reasoning process. Driven by this finding, we introduce a simple calibration approach, Attend to Your Own Thoughts (AYOT), where reasoning traces and final answers generated by the pre-trained high-precision target LLM on a proper set of calibration samples are used as the context input during the ternarization process, along with the corresponding questions. ScaleQ-1.58 is formed by simply integrating AYOT with CAT-Q, which demonstrates several scaling properties: (1) with only 4M calibration tokens, Qwen3-1.7B ternarized by ScaleQ-1.58 reaches over 90.52% of the performance of the prior best BitNet b1.58 2B4T averaged over 4 mathematics and coding tasks, and our ternary Qwen3-4B shows an absolute gain of 8.97%, while requiring 1,000,000x fewer calibration tokens for quantization; (2) ScaleQ-1.58 generalizes well to both dense and MoE architectures, with performance improving as model scale increases (up to 235B parameters); (3) ScaleQ-1.58 demonstrates strong generalization across tasks of varying difficulty levels, including mathematics, coding and scientific logic reasoning, as well as commonsense reasoning and basic language generation; (4) its performance continues to improve as the number of calibration tokens increases. Notably, AYOT also exhibits strong generalization ability across other quantization bit-widths. Code will be available at https://github.com/IntelChina-AI/BitTern.

cs.CL↗

arXiv2026

G-ReAct: Graph-Guided Deep Search via Structure-State Co-Evolution

Deep search has become a fundamental capability of large language models (LLMs) for solving open-domain complex tasks. However, existing approaches typically rely on linear sequential reasoning for both trajectory generation and inference, making it difficult to consistently preserve intermediate states and constraints throughout long-horizon multi-hop search. Consequently, they often suffer from context forgetting, search drift, and inefficient exploration. To address these limitations, we propose $\textbf{G-ReAct}$, a reasoning framework for deep search that organizes reasoning as $\textbf{state evolution over a fixed-topology query graph}$. The evolving graph state explicitly tracks search progress and guides subsequent decisions, transforming exploratory search driven by textual history into graph-guided reasoning under explicit constraints. G-ReAct supports both training and inference: it generates high-quality deep-search trajectories for supervised fine-tuning and provides structured guidance for inference-time search without additional fine-tuning. Experiments demonstrate that with only 1.9K generated trajectories for fine-tuning, Qwen3-30B-A3B-Thinking-2507 achieves $52.6\%$ accuracy on BrowseComp-ZH and $79.0\%$ on XBench, outperforming comparable open-source methods trained on substantially larger datasets, including RL-enhanced methods. Furthermore, when applied at inference time, G-ReAct consistently improves the performance of existing strong LLMs on deep-search tasks. We will publicly release all code and model weights.

cs.AI↗

arXiv2026

QR-Structured Thermal Triggers for Targeted Semantic Attacks on Infrared Vision-Language Models

Infrared vision-language models (IR-VLMs) extend thermal perception to open-vocabulary classification, image captioning, and visual question answering. However, their robustness to structured thermal perturbations and the stability of cross-modal semantic alignment remain insufficiently studied. We propose QR-Structured Thermal Triggers (QR-STT), a stealthy, training-free, black-box framework for targeted semantic steering of IR-VLMs. QR-STT preserves the functional regions of a QR pattern while optimizing its internal modules, each of which is assigned a cold, neutral, or hot thermal state. The framework jointly searches module topology and rendering parameters, including position, scale, rotation, intensity, blur, and roundness. A three-stage gradient-free procedure with greedy module-flip refinement efficiently handles the mixed discrete and continuous search space. The objective promotes alignment with an attacker-selected target, suppresses source-class evidence, and regularizes QR structure and visual similarity. Experiments on multiple CLIP-style encoders show that QR-STT consistently redirects image-text alignment toward chosen concepts while maintaining visual stealth. Perturbations optimized for classification also transfer to image captioning and VQA, causing target-consistent semantic drift in generated outputs. These results identify QR-structured thermal patterns as an interpretable attack surface for language-driven infrared perception and highlight the need for robustness evaluation against structured cross-task semantic attacks.

cs.CV↗

arXiv2026

Freq-RemoteVAR: Next-Frequency Autoregressive Modeling for Remote Sensing Change Detection

Remote sensing change detection aims to identify land-cover changes from bi-temporal images. Most existing methods follow a one-shot dense prediction paradigm, directly regressing a change mask from fused features. However, such approaches overlook the intrinsic frequency characteristics of change patterns. We propose Freq-RemoteVAR, a frequency autoregressive framework that reformulates change detection as a structured generation problem in the frequency domain. Instead of predicting the change mask in a single step, we introduce a next-frequency prediction paradigm, where change information is progressively generated from coarse to fine. We design a frequency-aware mask tokenization strategy that decomposes change supervision into multi-frequency token targets via Fourier transformation and quantization. We develop a Frequency VAR Transformer, which performs causal autoregressive modeling over frequency tokens. The model starts from learned mask queries and progressively predicts frequency-level tokens conditioned on previously generated tokens and bi-temporal image features, effectively capturing long-range dependencies across frequency scales. We introduce Scale-Aligned RoPE Cross Attention (SRCA) module, which aligns frequency-domain mask queries with spatial-domain bi-temporal features under a unified coordinate system, enhancing spatial-frequency consistency during generation. We propose a Change-quality Control module that adaptively modulates the generation process through dynamic normalization, attention biasing, and spatial offset adjustment, thereby suppressing pseudo-change responses and improving robustness. Extensive experiments on CDD, GZ-CD, and LEVIR-CD demonstrate that Freq-RemoteVAR consistently outperforms existing methods, particularly in challenging scenarios with complex appearance variations and noisy disturbances.

cs.CV↗

arXiv2026

Mi-Memory: A Lifecycle Memory Framework for Personal AI

Personal AI is moving beyond chat-only interaction toward continuous services that span phones, cars, homes, wearables, cameras, and tools. In this setting, memory cannot remain a cache of prior conversations. It should serve as a continuity and governance substrate: preserving durable user state, grounding answers in multimodal and device evidence, supporting correction and forgetting, bounding policy evolution, and remaining deployable under latency, cost, privacy, and edge-cloud constraints. This technical report presents Mi-Memory, a lifecycle memory framework for Personal AI organized around four roles: Structure, Expansion, Evolution, and Deployment. A shared audit contract links these roles through four recurring artifact families: typed evidence payloads preserve source identity and provenance, diagnostic traces localize evidence loss across the serving pipeline, strategy artifacts make memory-policy changes explicit, and gate/rollback records bound accepted evolution. MiMemory instantiates the roles through MemStack, MemSense/MemFuse, D$^{2}$ACCI/E$^{2}$MEND, and LiteMem. In controlled-reference Structure evaluations, MemStack reaches 93.59%, 57.24%, and 87.47% on LoCoMo, PersonaMem-V2, and LongMemEval, respectively; other tracks report module-level, preliminary/internal, transfer-feasibility, or design-only evidence with explicit boundaries. MiMemory is a step toward auditable, evidence-gated, and deployment-aware memory systems for Personal AI. Project homepage: https://darwin-agent.github.io/Mi-Memory/ .

cs.AI↗

arXiv2026

Locality-Aware Density Control for Efficient Gaussian-based Image Representation

2D Gaussian Splatting is an attractive direction for image representation due to its explicit formulation, fast rasterization, and favorable decoding efficiency. The representation quality of this paradigm depends on the proper allocation of Gaussian capacity to the demanding regions. However, existing methods fail to allocate Gaussian capacity efficiently during optimization: under-reconstructed content is often refined in a fragmented pixel-wise manner, while neighboring optimized Gaussians with similar attributes are redundantly retained. This inefficiency motivates the need for a density control framework that jointly addresses insufficient allocation in under-reconstructed regions and redundant allocation in over-reconstructed regions. Our key insight is that this framework should exploit two complementary forms of locality: the local continuity of reconstruction errors in image space for improved Gaussian allocation, and the local similarity of neighboring Gaussians in Gaussian space for redundant elimination. Based on this insight, we propose Locality-Aware Density Control (LocoADC), a plug-and-play framework that improves Gaussian capacity utilization through Region-wise Gaussian Densification (RGD) and Similarity-Driven Gaussian Merging (SDGM) strategies, together with a local color consistency constraint for more reliable merging. Extensive experiments on diverse datasets show that LocoADC consistently improves multiple baselines by enabling more effective local Gaussian allocation, including a 2.93 dB PSNR gain over GI on the CLIC dataset under the same 30k Gaussian budget. Code is available at: \textit{https://github.com/ChenJiaCong-1005/LocoADC}.

cs.CV↗

arXiv2026

FinResearchBench II: A Deep Research Benchmark with Consensus-Derived Gold Rubrics for Distinguishing Financial Report Quality

Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics. We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop. We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports. To justify removing human experts from rubric execution, we compare rubric judgments from three human experts with those from a three-LLM judge panel on a sampled subset, and show that LLM-based evaluation is sufficiently consistent with human evaluation to replace it for large-scale rubric screening, including 98.67\% label-level agreement on jointly unanimous items. We then derive consensus-derived gold rubrics through two filters: a strict consistency filter, which keeps a rubric only if the three LLM judges unanimously agree on every report under the same query, and a distinguishability filter, which keeps a rubric only if it assigns at least one majority-yes and at least one majority-no label across the evaluated systems. This process retains 3,687 consistency-passed rubrics, of which 2,600 remain distinguishable and form the final set of consensus-derived gold rubrics. Using this final rubric set, we obtain clearly differentiated rankings across 10 deep research systems, with item-level pass rates ranging from 58.58\% to 22.23\%. More broadly, because the pipeline removes human-expert execution from rubric generation and evaluation, it is naturally scalable for benchmark evaluation, automatic system comparison, and future studies of evaluation-driven system improvement.

cs.CL↗