arxiv arXiv cs.CL · 6h ago

WBCMor VQA: A Bilingual English-Urdu Hematology Visual Question Answering Benchmark

Researchers have introduced WBCMor VQA, a clinically validated bilingual benchmark for leukemia and normal white blood cell analysis in English and Urdu. This resource addresses the gap in multilingual healthcare technologies, particularly in regions like Pakistan where clinical documentation often mismatches patient communication languages. The dataset comprises 110,000 bilingual question-answer pairs annotated across 20,000 single-cell images of leukemic and normal white blood cells. To ensure linguistic consistency and clinical correctness, the benchmark utilizes morphology-aware annotations from the LeukemiaAttri and WBCAtt datasets alongside a domain-specific Urdu hematology dictionary. The study also highlights the limitations of existing English-centric vision-language resources in diverse healthcare environments. Baseline performance metrics were established by evaluating multiple open-source Vision Language Models on this new benchmark. This resource aims to facilitate the development of accessible AI systems for multilingual medical contexts.

arxiv arXiv cs.CL · 6h ago

Automatic Generation of Highlights for Academic Paper Via Prompt-based Learning

This study investigates prompt-based learning for the automatic generation of academic paper highlights to address the lack of labeled training data in existing supervised methods. The researchers designed task-specific prompt templates combined with paper abstracts as inputs for several language models, including locally deployed GPT-2 and T5, as well as ChatGPT accessed via API. Experiments conducted on three datasets demonstrated that ChatGPT with prompt templates achieved performance comparable to previous supervised methods without requiring task-specific training samples. When a small number of examples were added to the prompts, the model significantly outperformed state-of-the-art methods on two of the datasets. The analysis revealed that while ChatGPT possesses strong language modeling capabilities, its performance is highly sensitive to the specific information provided within the prompt. Case studies indicated that the generated highlights are generally coherent, informative, and closely resemble those written by authors. This approach does not rely on domain-specific training corpora, supporting downstream text mining and bibliometric research for papers lacking existing highlights.

arxiv arXiv cs.CL · 6h ago

Measuring Research Difficulty in NLP: An Inverted U-Shaped Relationship with Academic Impact

This study proposes a comprehensive evaluation system for measuring the difficulty of academic research, focusing on Natural Language Processing as a case study. The authors extract internal and external features from papers, including collaboration, content, and references, to compute multiple difficulty indicators. These indicators are weighted using the entropy weight method and summed to generate a final research difficulty score. Academic impact is quantified by citation frequency, while expert assessments validate the reliability of the measurement approach. Empirical results indicate that page count, reference count, and high-level institutional participation significantly correlate with academic impact. Crucially, the analysis reveals an inverted U-shaped relationship between research difficulty and impact. This suggests that moderately difficult research tends to achieve the highest level of academic influence.

arxiv arXiv cs.CL · 6h ago

Data-Driven Evolution of Library and Information Science Research Methods (1990-2022)

This study analyzes the influence of data-centric research on Library and Information Science by examining methodological evolution from 1990 to 2022. Researchers automatically extracted four key categories of data-driven entities from academic papers: algorithms and models, data resources, software and tools, and metrics. The analysis evaluates trends across three dimensions, including temporal characteristics, topic-specific evolution, and cross-method features. Findings identify data resources as the primary driver of methodological changes within the discipline. The research reveals a cyclical pattern characterized by emergence followed by stability or practical application in LIS methods. This perspective highlights how big data advancements have reshaped the field's technical landscape over three decades.

arxiv arXiv cs.CL · 6h ago

iLLaDA: An 8B Masked Diffusion Language Model with Fully Bidirectional Attention

The authors introduce iLLaDA, an 8B parameter masked diffusion language model trained from scratch using fully bidirectional attention. This approach contrasts with the predominant autoregressive factorization and causal attention used in modern large language models. The model's pre-training scaled to 12 trillion tokens, followed by supervised fine-tuning on a 25 billion-token instruction corpus for 12 epochs. iLLaDA maintains the masked diffusion objective throughout both training phases and employs variable-length generation for efficiency. It also introduces confidence-based scoring to enhance performance on multiple-choice evaluation tasks. Benchmark results show significant improvements over its predecessor, LLaDA, including gains of 21.6 points on BBH and 14.9 points on ARC-Challenge for the base model. The instruction-tuned variant achieved increases of 14.5 points on MATH and 16.5 points on HumanEval. Despite its non-autoregressive nature, iLLaDA remains competitive with Qwen2.5 7B across several metrics.

arxiv arXiv cs.CL · 6h ago

Hybrid-IR: Dual-Path Hybrid Retrieval with Iterative Reasoning for Complex Medical Question Answering

Large language models face challenges with hallucinations and outdated knowledge in biomedical applications, prompting the development of improved retrieval-augmented generation methods. Existing approaches often struggle with fragmented medical knowledge due to reliance on single retrieval paths and static strategies that hinder deep reasoning. To address these limitations, researchers introduced Hybrid-IR, a dual-path framework featuring an iterative retrieve-reason mechanism for complex medical question answering. This system integrates graph-based retrieval to explore structured knowledge alongside dense retrieval for fine-grained semantic matching. The model progressively refines its reasoning trajectory through an iterative loop between retrieval and reasoning steps. Experiments conducted on three widely used medical QA benchmarks demonstrate the effectiveness of this proposed approach.

arxiv arXiv cs.CL · 7h ago

Local Branch Routing: Efficient Trainable Test-Time Scaling for Language Models

The authors introduce Local Branch Routing (LBR), a token-level framework designed to improve language model reasoning through efficient test-time scaling. LBR expands a small local lookahead tree and forwards all sampled branches through the model, using a lightweight router to select the depth-1 subtree for commitment. This approach allows each token decision to utilize evidence from candidate local futures without incurring the computational costs of full solution-level search. The method employs a prune-shift-grow decoding process that preserves discrete branch identities and defines a tractable tree-trajectory likelihood. Consequently, LBR enables end-to-end reinforcement learning with verifiable rewards, jointly optimizing the base model and router under the same likelihood-ratio principle as discrete-token RLVR. Experimental results on synthetic hierarchical-planning tasks demonstrate that post-candidate hidden states provide useful routing evidence. Furthermore, benchmarks in mathematical reasoning show that LBR improves both Pass@1 and Pass@32 metrics compared to discrete chain-of-thought and other baselines.

arxiv arXiv cs.CL · 7h ago

Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents

Prior research on memory mechanisms in RAG-based conversational systems has primarily focused on storage and retrieval methods. This study investigates how memories with distinct functional roles influence response quality across varying contexts. The authors present a fine-grained taxonomy of conversational memory to classify retrieved items into specific role types. They also design a user-centric evaluation framework that simulates user perspectives to address limitations in reference-based assessments. Comparative experiments were conducted on long-term datasets using frontier large language models to analyze these effects. Results indicate that clarifying memory enhances factual accuracy and constraint awareness, leading to more correct and personalized responses. Conversely, irrelevant memory was found to reduce topic relevance and degrade constraint awareness capabilities. These findings demonstrate how different memory types can be leveraged to improve personalization in conversational agents.

arxiv arXiv cs.CL · 7h ago

Neural Machine Translation for Low-Resource Tangkhul-English

This study addresses low-resource machine translation for the Tangkhul-English language pair, focusing on a severely under-resourced Tibeto-Burman language with minimal prior NLP infrastructure. The authors present two systems: a primary model based on ByT5-large and a contrastive system using mT5-small, both fine-tuned on 38,336 parallel sentence pairs. Evaluation on a held-out test set of 3,856 sentences shows the ByT5-large system achieving a corpus BLEU score of 39.97 and a chrF++ score of 58.07. Additional metrics include a BERTScore F1 of 0.8104 and a COMET score of 0.7302 using the wmt22-comet-da model. The research highlights orthographic challenges related to Tangkhul's Latin-script diacritics as a specific technical hurdle. Furthermore, the training corpus exhibits domain bias, consisting primarily of biblical texts, stories, and conversational data. Future work aims to improve performance through data diversification and domain adaptation strategies.

arxiv arXiv cs.CL · 7h ago

Sarashina2.2-TTS: Tackling Kanji Polyphony in Japanese Speech Generation via Data Scaling and Targeted Data Synthesis

Sarashina2.2-TTS is a Japanese-centric LLM-based text-to-speech system designed to address the linguistic challenge of context-dependent kanji polyphony. The model scales training data to approximately 361k hours, utilizing a balanced mix of Japanese and English speech corpora. To specifically handle reading disambiguation, the authors implemented a targeted data augmentation pipeline covering all 2,136 Joyo regular-use kanji. Alongside the model release, the paper introduces the Joyo Kanji Yomi Benchmark, which includes 4,378 distinct readings for these characters. The authors also propose Kana-CER, a metric that evaluates pronunciation correctness by comparing synthesized speech against reference readings in kana space. Experimental results show that this targeted augmentation significantly improves reading accuracy and achieves state-of-the-art kanji-level performance. The system matches top baselines on general sentence-level pronunciation while delivering the highest speaker similarity in zero-shot synthesis scenarios. Furthermore, cross-lingual evaluations confirm that the balanced training approach ensures stable Japanese pronunciation regardless of the prompt language used.

arxiv arXiv cs.CL · 7h ago

Computational Stylometry of English Pali Canon Translations Across Pitakas

This study presents a computational stylometric analysis of the Tipitaka across all three Pitakas in English translation, extending previous work on the Sutta Pitaka. The corpus comprises 134,831 segments from Bhikkhu Sujato's Sutta Pitaka, Bhikkhu Brahmali's Vinaya Pitaka, I.B. Horner's 1938 Vinaya translation, three English translations of the Abhidhammattha Sangaha, and cross-tradition Vinaya texts. The authors compute Zipf rank-frequency distributions, MATTR-500 lexical diversity, numeral-word density, and vocabulary overlap metrics. Main findings indicate that all corpora show Zipf-consistent distributions with R-squared values above 0.989. The Sutta and Theravada Vinaya exhibit nearly identical lexical diversity scores of 0.399 and 0.400, while the Sangaha corpus is more diverse at 0.560. The Sangaha corpus also displays the highest numeral-word density at 3.26%, reflecting its systematic enumeration of categories. Additionally, the Mulasarvastivada Vinaya shares significant vocabulary overlap with the Theravada Vinaya, whereas two English translations of the same source share only 24.2% of their vocabulary.

arxiv arXiv cs.CL · 7h ago

Story Operators: Decomposing the Original to Sequel Transformation in Embedding Space

This study models literary transformations as geometric operations within a sentence-embedding space using all-mpnet-base-v2 vectors from the PG19 corpus. By calculating displacement vectors between original novels and their sequels, the author decomposes these changes along a content basis derived via PCA. Analysis of thirteen verified author pairs reveals a taxonomy of sequel types: formulaic, concentrated, and compositional. Formulaic transformations involve minimal rank changes, such as Doyle's Holmes collections with a norm of 0.12. Concentrated shifts are dominated by a single axis, exemplified by Alcott's Little Women to Little Men where 75% of the change occurs on one move. Compositional transformations involve many small axes, seen in works by Twain, Burroughs, and Nesbit. For Tom Sawyer to Huckleberry Finn, the dominant axis is structural, reflecting a shift from domesticity to picaresque adventure rather than surface themes like vernacular voice. The geometric findings are corroborated against Mark Twain's documented authorial intent in letters to Howells.

arxiv arXiv cs.CL · 7h ago

Survey of Toxicity Detection and Mitigation Strategies for Multilingual Language Models

This survey synthesizes research on toxicity detection and detoxification strategies specifically designed for multilingual large language models. It catalogs threat models that exploit linguistic variations such as code-switching, orthographic differences, and translation pivots to bypass safety alignments. The authors organize existing work into task formulations like toxic-to-neutral rewriting and classification, alongside various detection approaches including cross-lingual encoders and LLM-based detectors. Mitigation strategies are detailed across data filtering, supervised tuning, decoding-time steering, and the implementation of multilingual guardrails. The analysis highlights persistent challenges in the field, notably uneven language coverage and fragmented evaluation protocols. Furthermore, it addresses the complexity of culturally contingent definitions of harm and the risk that detoxification efforts may suppress legitimate dialectal or identity-related expression.

arxiv arXiv cs.CL · 7h ago

Introducing corpora Hlava Cor and Hlava AD: Human Label Variation in Coreference and Discourse Relations

Researchers have created two new corpora, Hlava Cor and Hlava AD, to explore human variation in understanding text coherence. These resources contain multiple annotations of Czech texts along with annotators' explanations for their choices. The first corpus, Hlava Cor, consists of 1,024 contexts annotated by three individuals to capture coreference identification differences. It covers pronouns, full noun phrases, and anaphoric adverbials across various text types and grammatical-semantic categories. The second corpus, Hlava AD, comprises 512 contexts annotated by five annotators focusing on discourse relations in attributive and non-attributive constructions. Both corpora achieve an inter-annotator agreement of approximately 60-65 percent. Analysis reveals that lower coreference agreement correlates with automatic model disagreement, indicating higher ambiguity. Annotator comments further highlight varying confidence levels and individual reading strategies.

arxiv arXiv cs.CL · 7h ago

Agent-Authored World Modeling Aligns Training with Decision Needs

The paper introduces Agent-Authored World Modeling (AAWM), a training procedure that addresses the limitations of standard world modeling objectives tied to next-observation prediction. This traditional approach often omits dynamics relevant to an agent's current decision because supervision depends on what a transition reveals rather than what is needed. AAWM constructs supervision directly from the policy's decision needs by having the agent identify necessary environmental understanding at each state. Relevant transition evidence is retrieved across trajectories and synthesized into training targets that capture these decision-oriented dynamics. This method aligns the learning objective with the specific information required before acting, rather than forcing the model to reconstruct the next observation. Experimental results validate AAWM's effectiveness across multiple environments and training settings. The findings demonstrate that decision-aware world-model targets provide a more effective learning signal than conventional next-observation prediction.

arxiv arXiv cs.CL · 7h ago

OscillaTTS: Adaptive Oscillatory Inductive Bias for Modeling Sharp Prosodic Dynamics in Diffusion-Based TTS

Diffusion-based text-to-speech models have improved speech quality but struggle with sharp prosodic transitions and rapid pitch variations. Existing decoders often use periodic nonlinearities like the Snake activation function, which lack adaptability for abrupt amplitude and frequency changes. To address this, the authors introduce OscillaTTS, a system featuring an adaptive oscillatory nonlinearity. This component enables controllable periodic modulation while ensuring signal stability via a linear bypass mechanism. The study investigates the role of oscillatory inductive bias within diffusion-based TTS decoders. Experiments conducted on the LJSpeech and Emotional Speech Dataset demonstrate consistent improvements in both objective and subjective evaluations. These results indicate that OscillaTTS effectively models expressive prosodic dynamics compared to prior methods.

arxiv arXiv cs.CL · 7h ago

Evaluating Japanese Dialect Robustness Across Speech and Text-based Large Language Models

This study investigates the dialectal robustness of large language models (LLMs) and speech language models (SLMs) using Japanese dialects as a test case. While LLM-based dialogue systems have advanced, dialectal variation remains a significant challenge, particularly for spoken input processing. The research defines robustness as the ratio of performance on dialectal versus standard inputs to enable fair comparisons across different model types. Experiments reveal that SLM robustness correlates directly with the robustness of their underlying text-based LLM counterparts. Additionally, the study finds that training with dialectal data and fine-tuning the speech encoder both serve to improve robustness in SLMs. These findings clarify how base LLM capabilities affect SLM performance and identify effective strategies for enhancing dialect comprehension.

github llama.cpp · 7h ago

Fix failed unit test cases for conv_3d in SYCL

The llama.cpp repository has addressed a specific issue regarding the SYCL backend. A pull request was submitted to fix the failed unit test cases associated with the conv_3d operation. This update targets the ggml-org/llama.cpp project on GitHub. The changes resolve errors that were previously preventing successful execution of these tests. This fix ensures better stability for users relying on SYCL-based hardware acceleration.

arxiv arXiv cs.CL · 7h ago

PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models

The authors introduce PolicyAlign, a framework designed to align large language models directly with natural-language safety policies rather than relying on costly supervision data. This approach addresses the mismatch between rapidly evolving safety requirements and conventional data-driven alignment methods. The process begins by synthesizing instructions that violate the specified policy, followed by on-policy self-distillation to internalize the desired behavior. To enhance training stability and data efficiency, the method incorporates Policy-Sensitive Filtering, which selects instructions inducing the largest behavioral shift. Experiments across multiple models demonstrate that PolicyAlign consistently improves safety metrics while maintaining low over-refusal rates and preserving general capabilities. The framework also generalizes effectively to specialized domains such as medical, legal, and financial safety scenarios. The code for this scalable alignment approach is released at https://github.com/Qwen-Applications/PolicyAlign.

arxiv arXiv cs.CL · 7h ago

Translation-Enhanced Speech Encoder Pre-training Improves Speech LLMs

Connecting a pre-trained speech encoder to a Large Language Model creates a structural misalignment because encoders often produce language-specific representations while LLMs operate in a unified, language-agnostic space. The authors argue that incorporating speech translation objectives into the pre-training process provides a principled mechanism to bridge this gap. Unlike monolingual transcription, translation forces the model to learn representations that are independent of specific languages. The study experimentally evaluates the impact of adding these translation objectives during speech encoder pre-training. Results demonstrate that this approach significantly improves cross-modal integration between the speech and text modalities. Consequently, models utilizing translation-enhanced pre-training achieve superior performance across various downstream Speech LLM tasks.