Multimodal Named Entity Recognition (MNER) uses supplementary images to resolve ambiguity in entity mentions in noisy, short-form social-media text. Prevailing approaches typically fuse text with visually aligned features, but such cues often lack sufficient discriminative power and may support conflicting entity-type hypotheses. To address this issue, we propose TCHV, a lightweight framework grounded in a type-conditioned hypothesis-verification perspective. By formulating each span-type pair as a distinct hypothesis during inference, TCHV predicts the latent feature evidence expected under each hypothesis and validates it against observed multimodal representations through bidirectional text-image matching, using a visual relevance gate to suppress unin- formative images. Verification scores serve as surrogate evidence statistics; TCHV improves over established non-VLLM multimodal baselines on Twitter-2015/2017, and TCHV† leads the compared VLLM-based methods on Twitter-2017. Extensions demonstrate applicability to grounded and fine-grained settings. Code: https://github.com/ hanfeiyu703/ImaginE.git.

Reasoning language models generate lengthy chain-of-thought (CoT) sequences whose key-value (KV) cache grows linearly and becomes a memory bottleneck during decoding. Existing compaction methods treat reasoning trajectories as flat token sequences and apply uniform compression, ignoring the hierarchical structure of CoT reasoning where different steps vary drastically in importance. We propose Thought-Aware Attention Matching (TAM), which exploits this structure through three mechanisms: (i)thought segmentation that decomposes the trajectory into reasoning blocks, (ii)adaptive budget allocation that assigns compression budget based on each segment’s importance and size, and (iii)pivotal token protection that preserves high-attention reasoning anchors. We prove that the allocation rule is optimal under a convex error model and that cumulative error under sequential compaction remains bounded. Experiments on AIME 2024 and MATH-500 with Qwen3-4B show that TAM improves accuracy over uniform compaction at the same memory footprint, with periodic compaction bounding peak memory to 3.1–3.2,GB (a 65% reduction) while maintaining competitive accuracy.

The democratization of LLMs has accelerated the generation and circulation of highly fluent disinformation, making traditional syntax-semantic verification increasingly insufficient. Such deception often combines persuasive rhetoric, emotional manipulation, and narrative role construction to shape readers’ interpretations. However, existing detectors typically emphasize isolated signals—such as syntax, external knowledge, persuasion, or affective cues—and thus struggle to capture multi-faceted manipulation or provide structured, taxonomy-grounded evidence. To address this gap, we present ExTax, a taxonomy-aligned framework for interpretable disinformation detection. ExTax unifies persuasive rhetoric, emotional manipulation, and narrative roles into a 17-dimensional taxonomic space comprising 6 persuasive-rhetoric strategies, 5 emotional-manipulation methods, and 6 narrative-role categories. It elicits attributes from multiple frontier LLMs, reconciles their disagreements through Entropy-driven Dynamic Label Smoothing, and fuses the resulting taxonomic representations with contextual encodings via Heterogeneous Multi-Head Attention. Across five cross-domain and cross-genre benchmarks, ExTax achieves an overall Macro F₁ of 0.8456, outperforming strong deep learning and LLM-based baselines. It also remains robust under severe genre imbalance, where the strongest deep baseline drops from 0.9454 to 0.6194 across genres. Our code is available at https://github.com/owlsan49/ExTax.

Nowadays, developing reliable DeepResearch-style long-form report generation remains challenging, as training and evaluation lack verifiable reward signals. Accordingly, rubric-based evaluation has become a common practice. However, existing approaches either rely on coarse, pre-defined rubrics that lack sufficient granularity or depend on manually constructed query-specific rubrics that are costly and difficult to scale. In this paper, we propose a pipeline to train preference-grounded query-specific rubric generators tailored for DeepResearch report generation. We first construct a dataset of DeepResearch-style queries annotated with human preferences over paired reports, and train rubric generators via reinforcement learning with a hybrid reward combining preference consistency, format validity, and LLM-based rubric evaluation. We evaluate the resulting rubric generators in two stages. First, on a held-out human-preference test set, the learned rubrics discriminate preferred from rejected reports more effectively than generic, prompted, or SFT-trained rubric alternatives. Second, when used as reward signals to train DeepResearch systems, our rubric generators yield substantial performance gains under both a simple single-agent ReAct framework and a complex multi-agent workflow on the DeepResearch Bench. Our code and model are available at https://huggingface.co/fdu-lcz/rubric_generator.

Knowledge graph embedding (KGE) models achieved state-of-the-art results on many knowledge graph tasks including link prediction and information retrieval. Despite the superior performance of KGE models in practice, we discover a deficiency in the expressiveness of some popular existing KGE models called Z-paradox. Motivated by the existence of Z-paradox, we propose a new KGE model called MQuinE that does not suffer from Z-paradox while preserves strong expressiveness to model various relation patterns including symmetric/asymmetric, inverse, 1-N/N-1/N-N, and composition relations with theoretical justification. Experiments on real-world knowledge bases indicate that Z-paradox indeed degrades the performance of existing KGE models, and can cause more than 20% accuracy drop on some challenging test samples. Our experiments further demonstrate that MQuinE can mitigate the negative impact of Z-paradox and outperform existing KGE models by a visible margin on link prediction tasks.