An analysis of model internals revealed a low-dimensional space with three interpretable dimensions matching a formal pre-specified framework for the geometry of legal concepts, showing how legal thought in LLMs may be organized and directed.
Peng Qian, Andrew Li, Sam S. Chen et al.· 0 citations
Agentic retrieval-augmented generation (RAG) enables language models to adapt retrieval based on previously retrieved evidence, but it remains unclear whether such adaptive orchestration consistently outperforms well-designed static pipelines. We conduct a controlled comparison of agentic and static RAG for Taiwanese h...
Kai-Hsin Chen, Wei-Yuan Chen, Xuan-Bo Chen et al.· 0 citations
Improving an existing methodology with a required control, an instruction-tuned model is compared against three RLVR agents trained from it, on four question-answering datasets, using only free-tier Kaggle GPUs.
It is argued that LLM-based AI systems answer political questions for hundreds of millions of people is a set of policies over whom to answer, what to say, and whether to engage at all, conditional on the topic and what the system knows about the user.
Joan C. Timoneda· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Multilingual code language models transfer skills across programming languages (PLs), but whether any PL occupies a privileged pivot position remains contested: geometric analyses point to C-family languages and Go, while behavioral evidence highlights Python. We revisit this question under controls for representationa...
AMUSER is presented, a multimodal framework which simulates user behaviour and automatically generates prioritised UX improvement recommendations from resulting data and suggests an asymmetric role of multimodality: visual access during simulation improves recommendations through richer traces, while providing visual i...
Anurupa Chowdhury, Bin Wu, Hossein A. Rahmani et al.· arXiv.org· 0 citations
Large language models (LLMs) increasingly mediate human decisions and communication, yet their behavioural regularities remain difficult to characterize systematically. We develop a cross-linguistic psychometric profiling framework and evaluate nine LLMs using seven psychological instruments, with five repeated adminis...
Non-technical stakeholders frequently cannot write the SQL needed to extract insights from operational databases. We built and evaluated a Text-to-SQL agent that closes this gap end to end: a six-node LangGraph StateGraph checks question relevance, fetches the live schema, generates PostgreSQL, validates it with a dry...
A methodology is implemented for evaluating large language models on sequential triage, the task of predicting a triage acuity label from a growing prefix of a nurse-patient conversation, which shows that the label at every checkpoint is anchored on the chief complaint exchanges, and prompting interventions fail to lif...
Dipankar Srirag, Hao-Kai Zhao, Ashutosh Kumar et al.· 0 citations
Semantic-Geometric Decoupled Routing is proposed, a training-free block routing framework that shifts semantic aggregation to the pre-RoPE space and reconstructs geometric bias with an offline structural prior and relative block distances and yields an explicit closed-form block routing score without token-level search...
This work demonstrates that traditional single- and multi-reference training paradigms are suboptimal for natural language generation tasks and provides empirical evidence that consolidating diverse references into a unified training signal is crucial for developing effective systems.
Maitreya Prafulla Chitale, Ketaki Shetye, Yash More et al.· 0 citations
Dialectal Arabic machine translation (MT) remains challenging despite recent progress in Arabic language technologies, particularly because effective translation requires modeling not only semantic content but also dialectal variation, conversational context, speaker and addressee characteristics, and sociolinguistic a...
Abdellah El Mekki, AbdelRahim A. Elmadany, Samar M. Magdy et al.· 1 citation