State Space Models (SSMs) such as Mamba have become a popular alternative to Transformer models, due to their reduced memory consumption and higher throughput at generation compared to their Attention-based counterparts. On the other hand, the community has built up a considerable body of knowledge on how to train Transformers, and many pretrained Transformer models are readily available. To facilitate the adoption of SSMs while leveraging existing pretrained Transformers, we aim to identify an effective recipe to distill an Attention-based model into a Mamba-like architecture. In prior work on cross-architecture distillation, however, it has been shown that a naïve distillation procedure from Transformers to Mamba fails to preserve the original teacher performance, a limitation often overcome with hybrid solutions combining Attention and SSM blocks. The key argument from our work is that, by equipping Mamba with a principled initialization, we can recover an overall better recipe for cross-architectural distillation. To this end, we propose a principled two-stage approach: first, we distill knowledge from a traditional Transformer into a linearized version of Attention, using an adaptation of the kernel trick. Then, we distill the linearized version into an adapted Mamba model that does not use any Attention block. Overall, the distilled Mamba model retains most of the original Pythia-1B Transformer performance, reaching a perplexity of 14.01 close to the teacher's 13.86. This largely carries over to downstream tasks, although some gaps appear on tasks relying more on in-context recall. To show the efficacy of our recipe, we conduct thorough ablations at 1B scale with 10B tokens varying sequence mixer architecture, scaling analysis on model sizes and total distillation tokens, and a sensitivity analysis on tokens allocation between stages.
Lakshmi Narayanan P, Narenkarthik K, Naren Kumar C
Claim verification is the task of deciding whether a source text supports a claim, contradicts it, or lacks sufficient evidence. Natural Language Inference (NLI) models can judge whether a single sentence supports or contradicts a claim, but they struggle when the evidence is spread across several sentences. Conversely, Large Language Models (LLMs) can reason across sentences, but may draw conclusions the text does not support. In this experimental study, we test whether a conservative fusion rule, labeling a claim Support or Contradict only when both models agree and defaulting to Not Enough Evidence otherwise, combines their strengths. We construct a benchmark of 2,035 claims from SciFact and CliniFact (858 Support, 152 Contradict, 1,025 Not Enough Evidence). We retrieve the top five most relevant sentences per claim using a hybrid of semantic and lexical (BM25) retrieval, which slightly outperforms either method alone. These retrieved sentences are then provided to every verification model. We then compare a DeBERTa-v3-base-mnli-fever-anli NLI baseline against Llama 3.1 8B, Qwen3 8B, and Mistral 7B. Qwen3 achieves the highest Macro-F1 (0.7937), followed by Llama 3.1 (0.7204) and Mistral (0.6388), while the NLI baseline reaches only 0.3157. Contrary to our hypothesis, fusion degrades performance, dropping Macro-F1 to 0.7392, 0.6730, and 0.5991, respectively. To understand this decline, we analyze agreements, disagreements, and common error patterns between the NLI and LLM verifiers. Our error analysis shows that NLI and LLM disagreements often result in correct LLM predictions being overturned.
Hongxuan Liu, Roman Bushuiev, Ivy Lightheart, Mrunali Manjrekar, Anton Bushuiev, Magdalena Lederbauer, Filip Jozefov, Yinkai Wang 等 15 位作者
Reliable benchmarking is critical for developing machine learning models for tandem mass spectrometry (MS/MS) based molecule discovery. Subtle issues in experimental design and model evaluation procedures can degrade the trustworthiness of such benchmarks and lead to erroneous conclusions. We conduct a thorough review of model evaluation issues in the recent MS/MS machine learning literature, using the standard MassSpecGym benchmark suite as a case study to illustrate the impact of these issues. We find evaluation issues in at least 17 of 26 papers reporting MassSpecGym benchmark results in the first year of its adoption. We isolate three classes of failures: (i) data leakage, (ii) shortcut learning, and (iii) implementation bugs and metric divergence. Through extensive experimentation and code replication, we quantify the impact of these issues and show how they corrupt the evaluation standards MassSpecGym was designed to enforce. We distill our findings into recommendations generalizable to MS/MS challenges, benchmarks, and custom evaluation setups. We also release MassSpecGym v1.5, an implementation of our recommendations in the MassSpecGym benchmarking suite which addresses the failure modes identified in this audit. MassSpecGym v1.5 is publicly available at https://github.com/pluskal-lab/MassSpecGym.
Self-supervised contrastive learning (SSCL) learns representations by contrasting augmented views of the same sample, but it can exhibit feature suppression, where dominant or easy-to-learn factors suppress other class-relevant signals, degrading the performance of representations, particularly in high-dimensional data with heterogeneous features. We introduce Fisher Contrastive Learning (FCL), which uses a regularized Fisher trace criterion to address feature suppression. FCL learns embeddings by maximizing the discriminative strength of Fisher discriminant functions, producing representations that preserve class-relevant information implied by the augmentation process. We prove that the representations learned by FCL could alleviate the feature suppression effects. We demonstrate FCL on simulated benchmarks designed to induce feature suppression and on a real-world biomedical application, pretraining on large white blood cell microscopy image dataset and transfer learning for diagnosis of acute promyelocytic leukemia in an external cohort. In all experiments, FCL outperforms the self-supervised baselines and yields more robust and discriminative representations.
Physics-informed holomorphic neural networks (PIHNNs) exploit complex-analytic representations to satisfy homogeneous PDEs by construction, but their extension to nonhomogeneous problems is nontrivial. We propose a hybrid physics-informed holomorphic neural network (HPIHNN) framework based on a decomposition of the solution into particular and homogeneous components. The particular component is computed by a spectral method on a simple extended domain, while the homogeneous component is approximated by a PIHNN trained only on the boundary with modified boundary data. The resulting formulation avoids meshing complex geometries and retains the dimensionality-reduction advantage of holomorphic representations. The method is applied to nonhomogeneous Poisson, Stokes, biharmonic, linear elasticity, and Helmholtz equations. Numerical experiments covering complex geometries, corner singularities, and nearly incompressible regimes demonstrate that the proposed method accurately resolves the corresponding solutions and, in the considered tests, generally achieves lower errors and shorter training times than standard PINNs. The results indicate that coupling spectral particular solutions with holomorphic neural approximations provides an efficient approach to nonhomogeneous PDEs on complex domains.
Multimodal Large Language Models (MLLMs) have achieved remarkable performance in Visually Rich Document Understanding (VRDU) tasks, but their capabilities are mainly evaluated on pristine, well-structured document images. We consider content restoration from shredded fragments, a challenging VRDU setting that requires integrating visual pattern recognition with semantic reasoning under significant content discontinuities. To facilitate systematic evaluation of complex VRDU tasks, we introduce ShredBench, a benchmark supported by an automated generation pipeline that renders fragmented documents directly from Markdown. The proposed pipeline ensures evaluation validity by allowing the flexible integration of latest or unseen textual sources to prevent training data contamination. ShredBench assesses four scenarios (English, Chinese, Code, Table) with three fragmentation granularities (8, 12, 16 pieces). Empirical evaluations on state-of-the-art MLLMs reveal a significant performance gap: The method is effective on intact documents; however, once the document is shredded, restoration becomes a significant challenge, with NED dropping sharply as fragmentation increases. Our findings highlight that current MLLMs lack the fine-grained cross-modal reasoning required to bridge visual discontinuities, identifying a critical gap in robust VRDU research.
Each new generation of frontier model pushes the edge of what is possible. Inside a harness -- a loop, a shell, a filesystem, the ability to write and run its own code -- a model can take on increasingly difficult work. A growing body of work treats the coding-agent harness as infrastructure for knowledge work rather than a coding tool, and a new wave of benchmarks tests it there. Most of them are generative: build a financial model, a memo, a spreadsheet from a brief, judged by a rubric or by another model. Few test the expensive part of knowledge work, which is investigative. In audit, tax, due diligence, regulatory reporting, claims or any dispute over numbers, the answer already exists -- hidden in incomplete and inconsistent evidence spread across systems -- and has to be recovered and defended to the cent. We introduce kw-bench, a benchmark built from real engagements specialists described to us. Its first task family is a five-year VAT reconstruction across three generations of SAP -- legacy ECC, a partial S/4HANA migration and a consultant's redesign -- delivered as raw system extracts: page headers, local number formats, partial link tables, no documentation of how anything connects. Grading is deterministic, with no LLM judge. We ran four harnesses -- Claude Code, Codex, OpenCode and micro-cc -- on the same model, DeepSeek v4.1 Flash, for two hours. The two independent, model-agnostic harnesses delivered (OpenCode 0.542, micro-cc 0.528); the two lab harnesses, away from their home model, delivered nothing (0.00 each). What defeated them was inertia: keeping many weak signals alive across a long session. On a frontier model the task collapses. micro-cc and Claude Code on Claude Opus 5.5 scored 0.9986 and 1.000, in 22 and 25 minutes. A weeks-long engagement now costs minutes and a few dollars; below the frontier, the harness decides how far a model gets.
A multimodal system may begin inference while holding only some of its inputs and may acquire the rest at a cost. With adaptive acquisition, the policy determines which inputs are ultimately observed, so we state risk control conditional on that terminal input pattern. Conditional calibration typically assumes that the grouping map is fixed independently of the calibration sample, a condition that policy-induced grouping does not satisfy. We characterize when pattern-conditional risk control remains valid and give two finite-sample constructions: threshold-free routing with calibration applied at the terminal pattern, and simultaneous validation of complete policy-pattern pairs, which lets calibration data select the deployed policy. A counterexample shows that validity proved for a calibration-independent grouping map need not transfer once the policy makes the terminal group calibration-dependent. We call the resulting method RouteCert. On a clinical electrocardiogram task with a staged, cost-ordered lead protocol, the deployed policy answers 71.2% of held-out patients, with an observed 7.4% disagreement with the cardiologist's diagnosis at 48.8% of the prespecified ordinal cost of acquiring each stage, and all three acquisition stages are validated separately. On masked multimodal benchmarks, validating pointwise at each terminal pattern holds observed worst-pattern selective risk, measured against the full-information reference decision rather than the true label, at 0.034, where a pooled design reaches 0.145 against a 0.10 cap, at a comparable answered fraction (0.350 vs 0.342); under the budget-matched simultaneous comparison, the answered fraction falls to 0.305.
Zijing Shi, Meng Fang University of Liverpool and Eindhoven University of Technology, Ling Chen University of Technology Sydney
As autonomous web agents are increasingly deployed to perform real-world tasks, ensuring their safety has become a critical concern. In this work, we study web agent behavior under realistic deceptive interfaces in the e-commerce domain. We introduce WebDecept, a lightweight and configurable plugin framework that enables controlled injection of deceptive interface patterns into existing web environments. Using WebDecept, we instantiate seven deceptive patterns commonly observed on the open web, including targeted advertisements, domain redirection, and shopping manipulation. By injecting these patterns into the frontend during task execution, we perform controlled evaluation of multiple multimodal web agents. Our results show that current web agents are highly susceptible to multiple classes of deceptive interfaces, and that prompt-based constraints are often insufficient to mitigate these failures. We further analyze how the design choices of deceptive patterns influence the success of such manipulations. These findings highlight safety challenges that should be addressed as web agents are scaled toward real-world deployment.
尚无 AI 解读，展示原文摘要
cs.CL · cs.CV
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
AI 速览：提出人口动态基础模型（PDFM），通过图神经网络融合多源地理数据生成嵌入，在27个地理空间任务上取得领先表现。
Supporting the health and well-being of dynamic populations around the world requires governmental agencies, organizations and researchers to understand and reason over complex relationships between human behavior and local contexts in order to identify high-risk groups and strategically allocate limited resources. Traditional approaches to these classes of problems often entail developing manually curated, task-specific features and models to represent human behavior and the natural and built environment, which can be challenging to adapt to new, or even, related tasks. To address this, we introduce a Population Dynamics Foundation Model (PDFM) that aims to capture the relationships between diverse data modalities and is applicable to a broad range of geospatial tasks. We first construct a geo-indexed dataset for postal codes and counties across the United States, capturing rich aggregated information on human behavior from maps, busyness, and aggregated search trends, and environmental factors such as weather and air quality. We then model this data and the complex relationships between locations using a graph neural network, producing embeddings that can be adapted to a wide range of downstream tasks using relatively simple models. We evaluate the effectiveness of our approach by benchmarking it on 27 downstream tasks spanning three distinct domains: health indicators, socioeconomic factors, and environmental measurements. The approach achieves state-of-the-art performance on all 27 geospatial interpolation tasks, and on 25 out of the 27 extrapolation and super-resolution tasks. We combined the PDFM with a state-of-the-art forecasting foundation model, TimesFM, to predict unemployment and poverty, achieving performance that surpasses fully supervised forecasting. The full set of embeddings and sample code are publicly available for researchers.
Kim, Ye-Chan, Cha, SeungJu, Kim, Si-Woo, Jeon, Minju, Kim, Hyungee, Kim, Dong-Jin
Weakly-Supervised Dense Video Captioning aims to localize and describe events in videos trained only on caption annotations, without temporal boundaries. Prior work introduced an implicit supervision paradigm based on Gaussian masking and complementary captioning. However, existing method focus merely on generating non-overlapping masks without considering their semantic relationship to corresponding events, resulting in simplistic, uniformly distributed masks that fail to capture semantically meaningful regions. Moreover, relying solely on ground-truth captions leads to sub-optimal performance due to the inherent sparsity of existing datasets. In this work, we propose SAIL, which constructs semantically-aware masks through cross-modal alignment. Our similarity-aware training objective guides masks to emphasize video regions with high similarity to their corresponding event captions. Furthermore, to guide more accurate mask generation under sparse annotation settings, we introduce an LLM-based augmentation strategy that generates synthetic captions to provide additional alignment signals. These synthetic captions are incorporated through an inter-mask mechanism, providing auxiliary guidance for precise temporal localization without degrading the main objective. Experiments on ActivityNet Captions and YouCook2 demonstrate state-of-the-art performance on both captioning and localization metrics.
尚无 AI 解读，展示原文摘要
cs.CV · cs.CL
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表