Classical and neural NLP: translation, question answering, tokenization and the evaluation of language understanding.
arXiv:2609.13869v1 Announce Type: cross Abstract: Automatic sentence function identification is important for many downstream natural language processing (NLP) applications such as dialogue systems,...
arXiv:2609.14245v1 Announce Type: cross Abstract: Context compression reduces generator input in retrieval-augmented generation, but answer quality alone does not characterize citation attribution. W...
The paper proposes a modular tokenizer framework for multilingual large language models, allowing the creation of language‑specific subtokenizers that match monolingual compression quality. It introduces a pretraining strategy that samples these subtokenizers to limit predictions to relevant vocabularies, enabling efficient training and inference. This approach reduces memory usage and speeds up inference without compromising performance.
arXiv:2609.13942v1 Announce Type: cross Abstract: The CRITICS project addresses science accessibility and literacy by converging advanced Machine Translation (MT) based on Large Language Models (LLMs...
arXiv:2609.14529v1 Announce Type: cross Abstract: Rigorous empirical validation of quantum machine learning on natural language tasks remains scarce. We evaluate a 10-qubit hybrid quantum-classical v...
The paper presents a system for the MedReason 2026 challenge that tackles both multiple‑choice and open‑ended medical visual question answering using offline, containerized inference. Key findings include that comparing answer semantics rather than labels boosts retrieval‑only accuracy from 20.0 % to 57.5 % on a 200‑case holdout, and that varying the number of in‑prompt retrieved examples has minimal impact on final accuracy (93.5 %–94.0 %). The final system achieves 94.0 % MCQ accuracy on the development set and 93.20 % on the official pre‑evaluation, far surpassing the off‑the‑shelf baseline. "whyItMatters":"The results demonstrate that semantic‑aware retrieval and careful adapter tuning can dramatically improve medical VQA performance, offering a practical approach for high‑accuracy, offline inference in clinical settings."
arXiv:2609.14523v1 Announce Type: new Abstract: Change visual question answering (Change VQA) requires understanding semantic changes across bi-temporal remote sensing images. Although vision languag...
FLoKD is an adaptive knowledge‑distillation framework designed for federated fine‑tuning of low‑rank LLMs over wireless networks. It transmits intermediate LoRA activations instead of full parameters or token‑level logits, and uses transformer block importance scoring plus dataset selection to reduce communication. Experiments on WikiText‑103, PTB, and Dialog show a 50‑65% reduction in communication while maintaining competitive perplexity.
FedV-KGQA addresses multi‑hop question answering over vertically partitioned knowledge graphs where each silo holds disjoint relation types. The system trains local embeddings, concatenates silo‑specific entity views, anchors questions at a topic entity, and ranks candidates without sharing raw triples. Experiments show federated fusion nearly matches centralized accuracy, that anchoring and enrichment are more critical than embedding choice, and that the cheapest encoder depends on target accuracy.
arXiv:2609.14448v1 Announce Type: cross Abstract: Hallucination in large language models reduces their reliability and slows adoption. Various white-box studies have used internal representations to...
The paper introduces ProNMT, a reward-guided iterative self‑training approach that balances global translation quality with pronoun‑specific feedback for context‑aware machine translation. ProNMT samples candidate translations, scores them using reference‑free quality estimation and a pronoun label derived from references, and fine‑tunes on the highest‑scoring candidate. Experiments on English–German Europarl and English–French News Commentary show that ProNMT outperforms standard context‑aware fine‑tuning on BLEU and COMET, while ablations reveal that pronoun‑only feedback can harm overall quality and that confidence‑weighted feedback outperforms hard binary feedback.
arXiv:2609.15224v1 Announce Type: new Abstract: We present LOVER, a \underline{L}ong to sh\underline{O}rt \underline{V}ideo \underline{E}vidence \underline{R}einforced model for grounded question ans...
arXiv:2609.14422v1 Announce Type: new Abstract: Agentic Network Operations (NetOps) are an emerging paradigm promising to enable workload-aware, self-adjustable, and reliable autonomous networks. Whi...
arXiv:2609.14825v1 Announce Type: cross Abstract: Large language models (LLMs) are often deemed unsafe for clinical question answering because of their tendency to hallucinate. Retrieval augmentation...
arXiv:2609.13815v1 Announce Type: new Abstract: Text-centric Visual Question Answering (VQA) requires reading and reasoning over text embedded in images, a task made substantially harder when images...
arXiv:2609.15478v1 Announce Type: new Abstract: Multimodal agents can create complex videos in software such as Blender by coding without relying on diffusion models. Yet video understanding benchmar...
MoVT is a new framework for text‑to‑motion generation that uses a cross‑modal augmented motion tokenizer to project 3D motion tokens into 2D, enriching the motion codebook with real‑world video patterns. The enriched tokens are mapped back to 3D, creating aligned 3D and 2D codebooks that better capture intricate motions. These codebooks feed a generative masked transformer, which predicts masked motion tokens in a modality‑agnostic way, allowing text‑index pairs from the 2D codebook and annotated videos to further improve generation quality. Empirical tests show MoVT outperforms previous state‑of‑the‑art methods on several key metrics.
arXiv:2609.13268v1 Announce Type: new Abstract: Annual reports, diligence packs, and infographic dashboards bury numbers in page images: axes, cell grids, and footnotes that OCR pipelines flatten and...
arXiv:2609.13288v1 Announce Type: new Abstract: Video-language models can answer multiple-choice questions with high confidence yet be wrong. We study whether answer-level reliability scores can be i...
The study investigates how different tokenization methods affect ECG Transformer models by comparing eight strategies across four backbone architectures on the CPSC2018 classification task. Physiology-aware tokenizations such as Median-beat and HeartLang achieve higher mean macro-AUCs (0.893 and 0.889) than point-wise and patch-wise approaches (0.822 and 0.824), while also reducing sequence length and training memory usage. Combining the two physiology-aware representations further improves macro-AUC by 8.2%.