Classical and neural NLP: translation, question answering, tokenization and the evaluation of language understanding.
arXiv:2609.21387v1 Announce Type: cross Abstract: Relexicalization is a pivotal technique in clinical NLP, as it facilitates robust masking of sensitive information while synthesizing datasets that r...
arXiv:2609.20849v1 Announce Type: new Abstract: Large Audio Language Models (LALMs) perform well on complex question answering but often show a reasoning gap, where explicit Chain-of-Thought (CoT) re...
arXiv:2609.22038v1 Announce Type: new Abstract: We introduce QuranicMMLU, a benchmark for evaluating generative AI on Quranic Arabic across multiple dimensions of linguistic complexity. Existing Qura...
arXiv:2601.06848v2 Announce Type: replace Abstract: Multimodal aspect-based sentiment analysis (MABSA) aims to identify aspect-level sentiments by jointly modeling textual and visual information, whi...
arXiv:2609.21844v1 Announce Type: new Abstract: Long transcripts are costly inputs for downstream NLP systems and often contain irrelevant context. We study query-conditioned topic localization: pred...
Previous work on Egyptian Arabic in NLP has focused largely on the prestigious Cairene Egyptian Arabic (CEA) dialect, resulting in a lack of representation for the less prestigious Sa'idi Egyptian Ara...
The paper investigates whether confidence signals from fine‑tuned large language models can improve extractive question answering that relies heavily on retrieval. Experiments on four 7‑9B model families show that retrieval alone recovers 92–99.8% of the best possible accuracy, leaving little room for confidence‑based routing or adaptation to help. The sequence‑likelihood confidence metric, even after recalibration or temperature scaling, fails to provide a statistically significant benefit across different correctness criteria and answer lengths, and the study ultimately offers a set of pre‑specified negatives with explicit dependencies as its main contribution.
The study investigates how individual components of a coding harness—planning, action space, and context management—affect autonomous coding agents’ performance. By fixing the execution loop and varying these components across 176 settings on SWE‑Bench Verified and Terminal‑Bench 2.1, the authors find that context management is most valuable when context windows are tight, staging rule‑based elision before LLM summarization yields the best efficiency, planning serves as an accuracy scaffold for weaker models and a cost saver for stronger ones, and predefined tools help models with limited bash skills while bash‑capable models benefit from a bash‑only interface. Trajectory‑level analysis shows that context management lengthens execution paths, planning alters where trajectories terminate, and the action space determines code granularity, offering a modular framework for future harness design.
The paper introduces probe guidance, a technique that leverages frozen internal states of a diffusion model to generate a guidance signal without requiring an extra forward pass during inference. This method improves continuous diffusion language models, achieving state‑of‑the‑art results on unconditional generation and enhancing performance on multiple‑choice question answering for a 1.7B model. The authors also use probes to analyze autoguidance, revealing that the weak model must originate from a low‑entropy training region to align dynamics with the strong model.
AntiGrounding is a visual action-selection framework that turns short robot trajectories into both executable motion plans and rendered prompts for vision‑language model evaluation. After filtering for feasibility, each trajectory is scored on safety, task alignment, efficiency, and physical plausibility using structured multi‑view visual question answering, and the best trajectories are refined and validated by a digital twin before real‑world execution. In eight real‑world manipulation tasks, the system achieved a 71.25% success rate with a single GPT‑6 Astra evaluator, outperforming baseline methods.
The paper introduces M2Tok, a Multi-head Multi-codebook Action Tokenizer that reduces reconstruction loss for continuous action signals by decomposing latent features into multiple heads and assigning independent codebooks to each. This design expands representational expressivity, leading to lower reconstruction error and higher success rates in Vision‑Language‑Action models evaluated on RoboTwin, Simpler‑Env, and zero‑shot real‑world tasks. Ablation studies confirm the effectiveness of both multi‑head and multi‑codebook mechanisms.
The paper presents a signal-theoretic machine learning framework to detect deceptive online job ads that facilitate forced labour. Using 464 verified cases from nine countries and 21 industries, the authors build multimodal models that combine computer vision, natural language processing, and semantic embeddings, achieving ROC‑AUC scores between 0.87 and 0.97. SHAP analysis identifies text quality, risk language, and visual features as key discriminators, and the authors deliver a proof‑of‑concept decision support system that outputs interpretable risk scores for practitioners.
Relational BabyLM is a decoder‑only Transformer that replaces standard self‑attention with a Dual Attention Transformer (DAT) to separate object‑level lexical features from structural/relational information. The model incorporates a Next‑Latent Prediction objective to compress history into a dense belief state and introduces a RoPE‑based symbol‑retrieval mechanism. On the BabyLM 2026 challenge, the best model ranks 6th overall and 3rd on the NLP‑task subset, outperforming GPT‑2 on most benchmarks and achieving the highest EWoK score among strict‑track entries.
The study analyzes 17,012 app‑store reviews for six major generative‑AI apps, using BERTopic and RoBERTa to uncover topics and sentiment. Negative sentiment is most common around advertising, authentication, server reliability, and subscription pricing, with significant differences across apps—Claude shows the highest negative sentiment yet a highly enthusiastic user base. The authors also note geopolitical and privacy concerns for DeepSeek and propose a Trust Friction Score to quantify trust and usability barriers.
DocAttriBench (DAB) is a large‑scale benchmark for fine‑grained, element‑level source attribution in Document Visual Question Answering (VQA). It introduces MAPPET, a Mask‑based Perplexity‑Derived Attribution method that uses document layout and language modeling to identify the most informative layout element for each answer. The benchmark contains 237k documents and 296k question‑answer pairs with element‑level grounding, and it evaluates multimodal LLMs on answer accuracy, attribution accuracy, and overall answer quality, revealing that even strong models often fail to localize supporting elements.
E-AVI is a new framework for automated video interview assessment that combines verbal, acoustic, and visual data. It extracts timestamped multimodal evidence and uses dimension‑conditioned attention with source‑level embeddings to score candidates. The system also provides a shared evidence pool for natural‑language feedback and follow‑up question answering, outperforming multimodal baselines on RecruitView and a private hospitality dataset.
The study investigates how document segmentation and chunk representation affect retrieval-augmented generation (RAG) for chemistry texts. Using the ChemQuests corpus, the authors benchmark 41 embedding models and evaluate them across five chunking strategies, seven chunk sizes, and various overlap settings. They find that embedding choice has the largest impact, with models like E5, BGE, and Nomic performing best, and recommend medium-to-large chunks with fixed-token, recursive-token, or hierarchical-section chunking and low overlap for effective chemistry-aware RAG.
VisKG‑LM proposes compiling retrieved knowledge graph subgraphs into static visual memories rather than re‑encoding them during each inference step. The method serializes each subgraph as Relation‑Labeled Paths, renders them as images that preserve the graph’s branching structure, and caches these images for reuse. At inference, a language model processes the question and candidate text first, then consults the cached visual memory only at its final layer, yielding improved performance on CommonsenseQA, OpenBookQA, and MedQA‑USMLE compared to both text‑only baselines and a large vision‑language model.
The article "The Life of a Token: from Words to Bits on the Wire" explores how large language models convert text into network traffic during training. It traces the transformation from words to tokens, then to vectors, and finally to binary streams that traverse high‑performance computing systems. Using Dante’s Divine Comedy as a case study, the tutorial examines how tokenization, embeddings, and parallelization affect the volume, structure, and timing of data exchanged across the network, and provides analytical traffic models and numerical examples to clarify the communication demands of LLM training.
The paper investigates how large‑language‑model (LLM) based AI agents mix latency, local resource usage, and container bottlenecks when processing user requests that involve remote LLM calls and local tool execution. By measuring three representative tasks—retrieval‑augmented question answering, web search, and software coding—the authors show that agents exhibit diverse resource dynamics, with concurrent requests revealing task‑specific bottlenecks in CPU, disk I/O, and memory. Leveraging these insights, they propose CPU‑aware tool admission and task‑aware CPU allocation, achieving up to a 5.4× speed‑up for CPU‑sensitive tasks and a 32% reduction in average latency across multiple tasks.