Classical and neural NLP: translation, question answering, tokenization and the evaluation of language understanding.
arXiv:2609.00241v1 Announce Type: new Abstract: Long documents often distribute important information across extensive narrative passages and multiple tables, making faithful summarization particular...
arXiv:2609.00832v1 Announce Type: new Abstract: The exponential growth of scientific publications calls for automatic Information Extraction (IE) systems to support knowledge discovery. In this conte...
arXiv:2609.01604v1 Announce Type: cross Abstract: LLM-based evaluators of natural language generation (NLG) quality are widely deployed as scoring tools and as automated training signals, yet the int...
arXiv:2608.12018v2 Announce Type: replace Abstract: Neural Machine Translation (NMT) and Large Language Models (LLMs) excel at cross-lingual tasks but often fail to capture intra-lingual morphologica...
EDRAC is the first large‑scale benchmark for dialectal Arabic machine reading comprehension and generative question answering, covering five major dialects—Egyptian, Moroccan, Emirati, Syrian, and Saudi. It contains 499 passages from naturally spoken interactions and 4,977 QA pairs produced via a human–LLM collaborative pipeline. The benchmark evaluates Arabic‑centric and multilingual large language models, revealing gaps between semantic answer quality and dialectal fidelity and underscoring limitations of current evaluation metrics for dialectal Arabic generation.
LatentPress compresses conversational histories and long documents into continuous memory tokens that a frozen decoder can read directly, eliminating the need for text reconstruction at inference. The method achieves 4–16× compression with only a small adapter (0.1% of the decoder’s parameters) and outperforms text summaries and OCR-based compression on LongMemEval and LongBench-QA benchmarks. Writing and reading are significantly faster than traditional text summarization or OCR reconstruction, demonstrating a practical machine-facing context interface beyond text and vision.
arXiv:2609.01200v1 Announce Type: new Abstract: When the visual encoder and the language decoder of a vision-language model (VLM) run on different compute nodes, the intermediate visual-token embeddi...
The paper introduces a semantic correctness taxonomy that categorizes open‑ended QA answers into eight ordered classes, distinguishing between correct, verbose, and hallucinated responses. It releases two datasets—CAP‑Correctness and CAP‑Statements—to support benchmark evaluation and NLI‑based training. The authors also propose CAP (Context‑Aware Precision), a reference‑based metric that scores question‑conditioned statements via bidirectional NLI and demonstrates superior performance under a monotonicity protocol.
arXiv:2609.01564v1 Announce Type: cross Abstract: Large language models (LLMs) struggle to classify text into taxonomies with many semantically similar labels, as the distinctions are domain-specific...
arXiv:2508.11857v3 Announce Type: replace-cross Abstract: Tokenization remains a persistent bottleneck in language modeling, especially when vocabulary learning is limited by whitespace boundaries. W...
Qwen-Drive-1.0 is a vision‑language foundation model tailored for autonomous driving that builds on a pretrained VLM architecture. It incorporates a bird’s‑eye‑view perception head for 3D object detection, semantic occupancy prediction, and BEV map segmentation, and a Planning Expert that generates future ego trajectories from shared representations. Experiments show strong 3D perception, driving scene understanding, and competitive motion‑planning performance while largely preserving general vision‑language capabilities.
The study investigates whether attention weights in Vision‑Language Models (VLMs) accurately reflect model reasoning for visual inputs. Using causal perturbation analysis, it identifies three distinct processing modes—Faithful‑Sufficient, Faithful‑Distributed, and Non‑Focal—indicating heterogeneous visual attention faithfulness. The research also shows that human‑annotated ground‑truth regions align with model attention in only about 60% of cases, highlighting a systematic divergence between model visual reliance and human intuition across VQA, document, and chart tasks.
The paper introduces a method for generating multilingual reasoning traces for medical question answering using large language models. It creates 500,000 reasoning traces in English, Italian, and Spanish by retrieving medical information from Wikipedia and applies them to MedQA and MedMCQA datasets extended into Italian and Spanish. The approach improves performance in both few‑shot in‑context learning and supervised fine‑tuning, achieving state‑of‑the‑art results for 8B‑parameter LLMs and releasing all resources for further research.
arXiv:2609.01356v1 Announce Type: new Abstract: Multilingual large language models (mLLMs) achieve strong performance in machine translation, yet our understanding of the mechanisms by which they tra...
arXiv:2510.26614v2 Announce Type: replace Abstract: We propose tokenization of events and present a tokenizer, Spiking Patches, specifically designed for event cameras. Given a stream of asynchronous...
The paper introduces SciGram, a large-scale dataset of 194K scientific diagrams paired with 1.4M visual instructions generated through a terminology‑grounded pipeline that extracts domain concepts, synthesizes facts, and retrieves relevant diagrams. Models fine‑tuned on SciGram show significant gains on diagram‑centric benchmarks such as TQA, ScienceQA, and AI2D, and when combined with existing models like LLaVA OneVision, set new state‑of‑the‑art performance. The authors release both the dataset and trained models to support further research in scientific diagram understanding.
DiscoTrace is a method that identifies rhetorical strategies used by answerers to information‑seeking questions by representing answers as sequences of question‑related discourse acts paired with interpretations of the original question, annotated on top of rhetorical structure theory parses. When applied to answers from nine different communities, DiscoTrace reveals that these communities exhibit diverse preferences for answer construction, whereas large language models (LLMs) lack such rhetorical diversity even when prompted to follow specific community guidelines. Additionally, LLMs tend to adopt a breadth‑oriented approach, addressing interpretations of questions that human answerers often ignore, highlighting a systematic difference in how LLMs and humans respond to information needs.
arXiv:2609.00588v1 Announce Type: new Abstract: Reranking methods, such as Minimum Bayes Risk (MBR) decoding and Quality Estimation (QE) reranking, are widely used in modern neural machine translatio...
arXiv:2602.09691v2 Announce Type: replace-cross Abstract: Knowledge distillation (KD) is a technique to compress a larger teacher system into a smaller student. In machine translation, KD is commonly...
KGFR introduces a Knowledge Graph Foundation Retriever that collaborates with large language models to enhance knowledge‑intensive question answering. By encoding relations with LLM‑generated descriptions and initializing entities from question roles, KGFR enables zero‑shot generalization to unseen knowledge graphs. Its Asymmetric Progressive Propagation technique efficiently handles large graphs, while a controllable reasoning loop allows the LLM to request candidate answers, supporting facts, and reasoning paths.