AI & Machine Learning · Computer Science & AI
Computation & Language / NLP
The newest papers in Computation & Language / NLP that ReadSearch has read, newest first. Each links to its own page, with its authors, date, where it was published and the original.
The newest 5,000 papers, from 17 September 2026 to 5 November 2026: 5,000 arXiv preprints.
Get the few papers worth your time in Computation & Language / NLP every morning.
Follow Computation & Language / NLP and ReadSearch reads its new papers each day, then brings you the few worth reading, each with its key finding in a sentence.
Newest papers, page 87
SZED at ArGuard Shared Tasks: Multi-Task Label-Attention and Cross-Modal Fusion for Arabic Hateful Meme Detection
We present SZED's submission to ArGuard 2026 Track A: Arabic hateful meme detection.Transformers Learn Two Competing Sorting Algorithms, and Only One Length-Generalizes
We show that two-layer transformers trained to near-zero loss on a sorting task converge to one of two mechanistically distinct solution classes depending on their initialization.Sharp Degree Bounds for Putinar Certificates in Terms of the Positivity Margin via Analytic Sums of Squares
We establish sharp degree bounds for Putinar sum-of-squares (SOS) certificates in terms of the positivity margin $δ$.How Far Can the Gemma 4 Speech Encoder Go Without Changing Its Weights? Saudi and Gulf Arabic Speech Recognition with a Frozen and an Adapter-Augmented Encoder
We study the 305M-parameter audio encoder shared by Gemma 4 and EmbeddingGemma 2 (identical except for its final projection) for Saudi and Gulf Arabic speech recognition, without changing any of its…What Matters for NVFP4 Training? A Comparative Study of Low-Precision Pretraining Recipes
Training large language models directly in 4-bit floating-point (FP) formats promises substantial improvements in throughput and energy efficiency.FlashMap: Flow Map Speculative Decoding
Speculative decoding accelerates large language model (LLM) inference by using a lightweight drafter to propose a block of tokens that the target model verifies in one parallel pass.Defending Against Dormant Poisoning Attacks Across Language and Multimodal Agents
Recent work shows that attackers can implant dormant malicious behaviors that remain hidden at release but emerge after benign fine-tuning.Long Listening Thoughts: Eliciting Open Auditory Reasoning with Deliberative Perception and Cognitive Refinement
We introduce Long Listening Thoughts (LLT), a framework for constructing post-training datasets that improves audio reasoning by pairing deliberate perceptual grounding with explicit cognitive…Ask Less, Ask Better: Allocating User Attention in Proactive LLM Agents with Interruption Budgets
LLM agents can ask clarifying questions to complete tasks, but answering each question demands users' time and effort.Toward a Universal Local Speech Feature Extractor Through Distillation
Self-supervised speech models commonly consist of a set of convolutional neural network (CNN) layers used as a local feature extractor, followed by transformer layers.When Opposite Premises Translate Identically: A Source-Conditioned Diagnostic for Multilingual Commonsense Items
Multilingual reasoning comparisons depend on translated premises preserving the relations in their English sources.Which Long Documents Matter? A Controlled Study of Data Selection for Context Extension
Natural long documents are widely used for context extension, but which selection pipelines are effective remains unclear.Pluralistic Alignment in Large Language Models: Objectives, Human References, and Preference Representations
Pluralistic alignment aims to represent meaningful variation in human preferences rather than collapse heterogeneous judgments into a single target.Ben-Misandry: A Bengali Dataset and Late-Fusion Transformer Framework for Detecting Anti-Male Gender Hate
Gender based hate speech research in Bengali has focused almost exclusively on misogyny, leaving anti male hate, or misandry, entirely unstudied for this language and creating a systematic blind…Beyond Recall Saturation: Scaling Hard-Negative Reranker Training for Industrial Legal Retrieval
Industrial retrieval systems often improve quality by enlarging embedding candidate pools or adopting complex training pipelines.Write Once, Check Many: Reusable Ledgers for Post-Hoc Policy Compliance Auditing of Agent Trajectories
Auditing LLM agents against policies is critical to ensure safety but requires rereading long trajectories once per policy.Modeling Discourse Profiles in Climate Misinformation Discourse
Prior work on climate misinformation detection typically classifies individual comments or claims using text alone.Vocal-Tract Acoustics, Aerodynamics, and Overlap-and-Add Synthesis in DYNARTmo
This paper presents the acoustic and aerodynamic extension of DYNARTmo, a dynamic articulatory model for visualizing and synthesizing speech production.Backend-Frontend Converter: Component-Based Conversion of .NET Computational Models to TypeScript
Computational models developed in .NET often need to run locally within web applications.ProbePO: Sparse Probes, Redistributed Credit for Critic-Free LLM Reinforcement Learning
Reinforcement learning from verifiable rewards assigns one terminal reward to an entire reasoning trajectory, obscuring how individual reasoning steps contributed to the outcome.