mlsanityData to 14 Oct 2026

ARMOR: Adaptive Retriever Optimization for Low-Resource Telecom Question Answering

arXiv 2606.29706 ↗First posted 29 Jun 2026cs.IR cs.AI cs.CLcited by 0

On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval

Abstract (arXiv)

Telecom question answering (QA) is a challenging setting for retrieval-augmented generation (RAG): evidence is fragmented across standards, papers, encyclopedic resources, and web documents, and answers often hinge on technical tables, equations, and specialized protocol language. In low-resource subdomains, generator fine-tuning can over-specialize and degrade general capability, making query-side retriever adaptation an attractive alternative. To this end, we ask whether a fixed-generator, query-adapted RAG system can outperform generator-side adaptation, and which retriever objectives best support that setting. We motivate retrieval, rather than generator fine-tuning, as the adaptation target through a capacity comparison: under bounded-parameter and soft-retrieval assumptions, query-encoder tuning can have a smaller estimation term than supervised fine-tuning when its effective dimension is smaller. We identify two particularly relevant objectives -- the latent-document RAG likelihood, which optimizes generation utility, and the InfoNCE contrastive objective, which improves semantic retrieval geometry -- and leverage them jointly through a retriever optimization method targeting downstream QA performance in the telecom domain. Specifically, we introduce ARMOR, Adaptive Regularized Mixture Optimization for Retrievers, which learns separate temperatures for the RAG retrieval distribution and InfoNCE softmax and regularizes the adapted query encoder toward the frozen base query encoder. Across telecom-specific retrieval and generative QA benchmarks, we show that ARMOR improves evidence retrieval and answer generation in several in-domain settings. Code is available at https://github.com/heshandevaka/ARMOR.git.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 33 of its 40 references that are in the map. First posted on arXiv 29 Jun 2026.

Builds on

33 of 40 references.

References · 33

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  3. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  4. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  5. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  6. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  7. 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,358
  8. 2021Improving language models by retrieving from trillions of tokenscited by 1,293
  9. 2020Gradient Surgery for Multi-Task Learningcited by 1,256
  10. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
  11. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,216
  12. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,210
21 more
  1. 2023Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!cited by 1,154
  2. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 952
  3. 2017GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networkscited by 834
  4. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 821
  5. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 792
  6. 2023In-Context Retrieval-Augmented Language Modelscited by 702
  7. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  8. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 573
  9. 2019Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networkscited by 566
  10. 2024ORPO: Monolithic Preference Optimization without Reference Modelcited by 562
  11. 2021Conflict-Averse Gradient Descent for Multi-task Learningcited by 459
  12. 2023Ragas: Automated Evaluation of Retrieval Augmented Generationcited by 418
  13. 2024RAFT: Adapting Language Model to Domain Specific RAGcited by 262
  14. 2023On Penalty-based Bilevel Gradient Descent Methodcited by 87
  15. 2023TeleQnA: A Benchmark Dataset to Assess Large Language Models Telecommunications Knowledgecited by 63
  16. 2025UFT: Unifying Supervised and Reinforcement Fine-Tuningcited by 59
  17. 2023Sequence Length Independent Norm-Based Generalization Bounds for Transformerscited by 29
  18. 2024Tele-LLMs: A Series of Specialized Large Language Models for Telecommunicationscited by 25
  19. 2022Mitigating Gradient Bias in Multi-objective Learning: A Provably Convergent Stochastic Approachcited by 14
  20. 2024Understanding Forgetting in LLM Supervised Fine-Tuning and Preference Learning -- A Convex Optimization Perspectivecited by 12
  21. 2024A Statistical Framework for Data-dependent Retrieval-Augmented Modelscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Leveraging Fine-Tuned Retrieval-Augmented Generation with Long-Context Support: For 3GPP Standards
  2. 2026Retriever Portfolios: A Principled Approach to Adaptive RAG
  3. 2024TeleOracle: Fine-Tuned Retrieval-Augmented Generation with Long-Context Support for Network
  4. 2025Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers
  5. 2026One Size Does Not Fit All! Dynamic Retriever and Generator Selection for RAG
  6. 2025Investigating the Robustness of Retrieval-Augmented Generation at the Query Level
  7. 2024Efficient In-Domain Question Answering for Resource-Constrained Environments
  8. 2025Towards Adaptive Memory-Based Optimization for Enhanced Retrieval-Augmented Generation