Multi-hop question answering (QA) requires systems to iteratively retrieve evidence and reason across multiple hops. While recent RAG and agentic methods report strong results, the underlying retrieval–reasoning process is often left implicit, making procedural choices hard to compare across model families. This survey takes the execution procedure as the unit of analysis and introduces a four-axis framework covering overall execution plan, index structure, next-step control strategies and triggers, and stop/continue criteria. Using this schema, we map representative multi-hop QA systems and synthesize reported ablations and empirical tendencies on standard benchmarks, including HotpotQA, 2WikiMultiHopQA, and MuSiQue. We highlight recurring trade-offs among effectiveness, efficiency, and evidence faithfulness, and conclude with open challenges for retrieval–reasoning agents, including structure-aware planning, transferable control policies, and robust stopping under distribution shift.

Retrieval-augmented generation (RAG) grounds large language models (LLMs) in external evidence, yet multi-hop pipelines still suffer from redundant sub-queries, shallow exploration, and premature or delayed stopping. We present EVO-RAG, a phase-aware framework that couples a lightweight two-stage curriculum (Discovery→Refinement) with seven step-level rewards and an in-episode time scheduler. The scheduler decays exploration incentives as evidence accumulates while increasing efficiency and correctness pressure as uncertainty shrinks. Beyond scalar rewards, we train a multi-head preference model and benchmark DPO, PPO, and GRPO under identical rollouts and curricula for a controlled comparison. Evaluated on HotpotQA, 2WikiMultiHopQA, MuSiQue, and Bamboogle with 8B-class backbones, EVO-RAG improves EM/F1 while reducing redundant hops. Ablations show that suppressing query overlap, rewarding controlled backtracking and justified refusal, and time-dynamic weighting are key to the accuracy–efficiency trade-off.

Large language models have achieved strong performance on medical reasoning benchmarks, yet their deployment in clinical settings demands rigorous verification to ensure factual accuracy. While reward models offer a scalable approach for reasoning trace verification, existing methods face two limitations: they produce only scalar reward values without explicit justification, and they rely on single-pass retrieval that precludes adaptive knowledge access as verification unfolds. We introduce Med-TIV, an agentic framework that addresses these limitations by training medical reward models to iteratively query external medical evidence during evaluation. Our approach combines tool-augmented verification with an iterative reinforcement learning paradigm that requires only trace-level supervision, alongside an adaptive curriculum mechanism that dynamically adjusts training data distribution. Across four medical reasoning benchmarks, Med-TIV achieves substantial gains over existing methods, improving MedQA accuracy by 23.5% and MedXpertQA by 32.0% relative to the base generator. Crucially, Med-TIV demonstrates an 8× reduction in sampling budget compared to prior reward model baselines. These findings establish that agentic medical reward modeling offers a principled path toward more reliable medical reasoning systems.

Biomedical retrieval-augmented generation (RAG) produces long-form answers that mix supported, under-evidenced, and contradicted claims, demanding fine-grained verification. We propose MedRAGChecker, a claim-level diagnostic framework that combines textual natural language inference (NLI) with a biomedical knowledge-graph (KG) compatibility signal. Its asymmetric selective-gating rule preserves the textual prediction for most claims and applies KG adjustment only to confident textual entailments assigned low plausibility by the KG back-end. We distill GPT-4.1 claim decompositions and NLI pseudo-labels into compact student models and evaluate three biomedical generators on PubMedQA, MedQuAD, LiveQA, and MedRedQA. The student-ensemble results show that selective BioPortal gating increases macro-averaged faithfulness from 60.8% to 64.4% and reduces the contradiction rate from 19.9% to 15.0%. On 172 annotations of 86 claims whose decisions change after fusion, agreement with human evidence judgments increases from 63.4% to 69.8%. A comparison with DRKG shows that fusion behavior varies substantially with KG coverage, entity linking, score construction, and calibration; the comparison therefore characterizes back-end effects rather than isolating a causal effect of coverage.

While reasoning capabilities typically emerge in large language models (LLMs) with tens of billions of parameters, recent research focuses on improving smaller open-source models through knowledge distillation (KD) from commercial LLMs. However, many of these studies rely solely on responses from a single LLM as the gold rationale, unlike the natural human learning process, which involves understanding both the correct answers and the reasons behind mistakes. In this paper, we introduce a novel Fault-Aware DistIllation via Peer-Review (FAIR) approach: 1) instead of merely obtaining rationales from teachers, our method asks teachers to identify and explain the student’s mistakes, providing customized instruction learning data; 2) we design a simulated peer-review process between teacher LLMs, and selects only the generated rationales above the acceptance threshold, which reduces the chance of teachers guessing correctly with flawed rationale, improving instructional data quality. Comprehensive experiments and analysis on mathematical, commonsense, and logical reasoning tasks demonstrate the effectiveness of our method. Our code is available at https://github.com/zhuochunli/Learn-from-Committee.

This paper introduces the RAG-RLRC-LaySum framework, designed to make complex biomedical research accessible to laymen through advanced Natural Language Processing (NLP) techniques. Our innovative Retrieval Augmentation Generation (RAG) solution, enhanced by a reranking method, utilizes multiple knowledge sources to ensure the precision and pertinence of lay summaries. Additionally, our Reinforcement Learning for Readability Control (RLRC) strategy improves readability, making scientific content comprehensible to non-specialists. Evaluations using the publicly accessible PLOS and eLife datasets show that our methods surpass Plain Gemini model, demonstrating a 20% increase in readability scores, a 15% improvement in ROUGE-2 relevance scores, and a 10% enhancement in factual accuracy. The RAG-RLRC-LaySum framework effectively democratizes scientific knowledge, enhancing public engagement with biomedical discoveries.