CORTEXA
← Browse
arxivcs.CLcs.AI2026-07-23

Data Quality over Capacity: Internalizing Documents into LoRA Adapters for Closed-Book QA

Joan Figuerola Hurtado

We study baking documents directly into the weights of a 4-bit Gemma-4-e4b model via LoRA, so a system can answer questions about a corpus closed-book: no retrieval and no context-window budget. Across roughly 100 training runs from single documents to a 99-document corpus, we find that once adapter capacity is adequate, training-data quality is the dominant lever on closed-book accuracy, outweighing LoRA rank, learning rate, and two alternative architectures combined; capacity itself is a hard gate below which no data intervention helps. A single curation pass (shortening gold answers to canonical 1-6 word spans and dropping trivia) moved closed-book accuracy from 57.7% to 85.7% on a 15-document corpus, a larger jump than any architectural change. We confirm a capacity trend (rank must grow with corpus size) entangled with a coupling between rank and learning rate that we initially misdiagnosed. On a 15-document slice we add a real retrieval baseline: the internalized adapter (84.2% recall) beats a BM25-RAG pipeline with a base reader (58.9%) and even a realistic gold-chunk oracle (65.6%) at lower latency. We report the full arc, including three misdiagnoses, as a case study in debugging LLM training empirically.

View free PDFSource page

Related papers

arxivcs.CLcs.AI2026-07-23

Artificial Epanorthosis: Why large language models overuse a classical rhetorical figure, and how to mitigate it

Federico Boggia

A rhetorical figure that Cicero and Quintilian catalogued two thousand years ago reappears, systematically, in the text of large language models: epanorthosis, the self-correction of the specimen «This is not a course. It is a journey of transformation». This essay argues that th…

View free PDFSource page
arxivcs.AIcs.CLcs.CVcs.RO2026-07-24

Zero-Shot Mission-Level Evaluation for Aerial MLLM Agents

Suman Navaratnarajah, Taehyoung Kim, Jona Ruthardt, Ishaan Bhimwal, Ryousuke Yamada, Yannik Blei, et al.

Multimodal Large Language Models (MLLMs) are emerging as core reasoning modules for embodied agents, yet it remains unclear how well general-purpose models can solve long-horizon embodied tasks from a single high-level instruction. We introduce MissionBench, a benchmark for missi…

View free PDFSource page
arxivcs.AIcs.CL2026-07-24

Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Mode

Nanbeige Lab, :, Chen Yang, Chengrui Huang, Fufeng Lan, Hanhui Chen, et al.

We present Nanbeige4.2-3B, a compact general agentic model with 3B non-embedding parameters. It delivers strong performance across code-agent, office-agent, and complex tool-use tasks while maintaining highly competitive reasoning capabilities in mathematics, coding, and science.…

View free PDFSource page
arxivcs.AIcs.CLcs.HCcs.MAcs.SE2026-07-23

HiMe: Real-Time Self-Hosted Personal Agent Platform for Health Insights with Wearable Devices

Wei Liu, Siya Qi, Linhai Zhang, Lorainne Tudor Car, Yulan He

Traditional approaches to wearable health signal analysis, such as smartwatches, are constrained by rigid analytical frameworks and limited personalisation. The emergence of LLM agents creates a new opportunity for Personal Health Agentic Analysis, where health insights can be ge…

View free PDFSource page
arxivcs.CLcs.AI2026-07-23

One More Turn, Less Regret: A Regret-Based Multi-Turn Benchmark for LLMs' Clarification Policies

Minh Ngoc Ta, My Anh Tran Nguyen, Duong D. Nguyen, Yuxia Wang, Preslav Nakov

Ambiguous user requests make clarification a sequential decision problem for conversational LLM assistants: they must decide whether to ask, what to ask, when to stop, and when to answer. We introduce RegretBench, a multi-turn benchmark that evaluates clarification as policy beha…

View free PDFSource page