A new arXiv preprint introduces PolicyMem, a method that uses geometric policy memory to support governance of large language models. The authors frame their work as a response to the limitations of current safeguard approaches, which they group into learning-based guards and a second paradigm. The paper is listed as a new submission in the cs.CL category.
A new arXiv preprint introduces RSIAgent, a multi-agent system that lets digital agents explore unfamiliar environments and iteratively improve themselves without any additional training. The approach is designed for settings where interfaces, tools, and failure patterns differ from what pretrained models have seen. The work appears under arXiv:2609.15364v1 in both cs.AI and cs.CL.
A research paper examines syntactic dependency trees drawn from 124 languages chosen for typological, genetic and geographic diversity. The author investigates the structural properties of these trees, noting that the organizing principles behind their topology are still not well understood. The work aims to identify regularities that hold across languages.
A new arXiv preprint in computational linguistics studies how arguments are structured when a direct proof of a claim is difficult, and how an alternative but related statement can be used instead. The authors apply large language models to compare argumentation and proof patterns across different text genres. The work falls within NLP research on reasoning and argument mining.
A arXiv paper examines where the concept of animacy is encoded inside large language models. The authors trace internal circuits tied to the animate/inanimate distinction, which involves verb-argument constraints and contextual cues beyond simple word-level features. The work is a revised version of a preprint in the cs.CL category.
A new arXiv paper in computational linguistics introduces MUSE, a story-generation engine that draws on narrative theory to steer how plot, character, and language choices fit together across planning, drafting, and revision. The authors frame story guidance as facing two bottlenecks, pointing to the quality of the guidance and how it is supplied. The abstract as posted is truncated, so full details of the method and results were not available.
This arXiv paper notes that intraoral scan datasets such as Bite2Text arrive already aligned in occlusion, which means key occlusal measurements can be calculated directly rather than inferred by a learned captioning model. The authors argue for deriving these quantities in closed form as a basis for automatically generating orthodontic reports. The work falls under computation and language research and has not been peer reviewed.
A newly posted arXiv study tackles a core limitation of decoding speech from non-invasive brain recordings: the neural signals are weak and noisy, making phoneme- or word-level reconstruction unreliable. Drawing on neuroscience evidence about how the brain encodes meaning, the authors propose recovering high-level semantic content as an intermediate step instead. The paper is cross-listed in the computational linguistics and machine learning categories.
A newly posted arXiv paper introduces CityPlanner, a sandbox agent that treats urban planning as an executable optimization task. The agent chooses among large sets of candidate actions while accounting for practical constraints like budget and the quality of public services. The work was listed in the AI category and cross-listed under computational linguistics.
A new arXiv paper argues that knowledge distillation breaks down when the student model is much smaller than its teacher, because imitating the teacher's logits locks the student into its sharpest output modes and harms generalization. The authors propose letting the large teacher guide the small student through prompts during reinforcement-learning fine-tuning instead of through gradient-based distillation. The work appears in the computational linguistics category on arXiv.
Researchers introduced MADS, a scalable framework that produces multi-turn persuasive conversations through agent self-play. The setup uses three coordinated agents, including user agents that role-play varied persona-driven behaviors, to generate diverse dialogue datasets. The paper appears on arXiv with cross-listings in artificial intelligence and computational linguistics.
A research paper on arXiv introduces a method for scoring text against evaluation rubrics using large language models, addressing how black-box judge models can apply identical criteria in inconsistent ways. The approach ties each score to concrete evidence drawn from the evaluated text, making the reasoning behind judgments easier to audit and reproduce. The work is cross-listed under arXiv categories for artificial intelligence, computational linguistics, and machine learning.
A computational linguistics paper investigates how readers actually experience AI-translated literary works, looking beyond surface accuracy to immersion and literary impact. The findings suggest that while machine translations convey the content acceptably, readers still favor human versions, indicating that common automatic evaluation metrics overlook qualities that matter most in literary reading.
A preprint cross-listed in arXiv's AI and computational linguistics categories examines how software could infer what a person is ultimately trying to achieve based on their routine computer activity. Leveraging recent progress in user modeling, the authors describe an approach in which AI systems and users jointly articulate life goals rather than the machine simply automating individual tasks. The v2 announcement indicates a revised version of the paper.
A new arXiv paper in computational linguistics examines how people's written narratives can be analyzed to support mental health assessment. The authors argue that existing approaches, from dictionary-based counting to neural methods, remain fragmented and overlook discourse-level structure. Their proposed multi-level narrative evaluation reportedly outperforms purely lexical features on related tasks.
An arXiv paper introduces s-Trace, a technique for measuring how much of their computational capacity large language models actually use on different inputs. The authors note that models with billions of parameters arranged in deep computational graphs may not fully exploit their capacity for every input. The updated paper is cross-listed under the AI, computational linguistics, and machine learning categories.
A new computational linguistics paper tests whether the emotional tone of U.S. lawmakers' communication is tied to actual legislative outcomes. Unlike earlier work that mainly charted how often and how politicians express emotion, the authors investigate whether those patterns carry practical consequences for lawmaking effectiveness.
A new arXiv paper examines why large language models tend to respond uniformly to different cultural groups even when context should call for differentiation. The authors combine mechanistic interpretability with a factorial experimental design on a cultural-appropriation benchmark, locating specific attention heads they term 'cultural binding heads' that appear tied to this behavior. The work is cross-listed on arXiv under machine learning, artificial intelligence, and computational linguistics.