PANINI-DR: Controlled Deep Research through Generative Semantic Workspaces

cs.CL, cs.AI, cs.IR

Submitted: 2026-10-10

Updated: 2026-10-16

Comments: Under review at ICLR 2027. 37 pages, 1 figure, 22 tables. M. Yigit Turali and Shreyas Rajesh contributed equally

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Open Deep Search: Democratizing Search with Open-source Reasoning Agents
• HyCE-RAG: Hypergraph Chain-of-Evidence Retrieval-Augmented Generation for Explainable Multi-hop Question Answering
• IterResearch: Rethinking Long-Horizon Agents with Interaction Scaling
• Prove2Me: An Open Collaborative Platform for Scaling Math Formalization
• BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
• R2VC: Modular Fact-Checking with Retrieval, Verification, and Confidence Calibration
• Memory-Aware and Uncertainty-Guided Retrieval for Multi-Hop Question Answering
• Fact, Fetch, and Reason: A Unified Evaluation of Retrieval-Augmented Generation
• LLM-as-a-Verifier: A General-Purpose Verification Framework
• CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning
• SentGraph: Hierarchical Sentence Graph for Multi-hop Retrieval-Augmented Question Answering
• Let's Verify Step by Step
• Track, Rank, Crack: Epistemic Working Memory Scales Multi-Hop Reasoning in Language Agents
• Omni-Decision: Evidence-Ledger Planning for Omni-Modal Agents
• Self-Refine: Iterative Refinement with Self-Feedback
• MA-RAG: Multi-Agent Retrieval-Augmented Generation via Collaborative Chain-of-Thought Reasoning
• MemGPT: Towards LLMs as Operating Systems
• Panini: Continual Learning in Token Space via Structured Memory
• Reasoning in Trees: Improving Retrieval-Augmented Generation for Multi-Hop Question Answering
• Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering