REVIEW 5 cited by
Retrieve, Summarize, Plan: Advancing Multi-hop Question Answering with an Iterative Approach
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Multi-hop question answering is a challenging task with distinct industrial relevance, and Retrieval-Augmented Generation (RAG) methods based on large language models (LLMs) have become a popular approach to tackle this task. Owing to the potential inability to retrieve all necessary information in a single iteration, a series of iterative RAG methods has been recently developed, showing significant performance improvements. However, existing methods still face two critical challenges: context overload resulting from multiple rounds of retrieval, and over-planning and repetitive planning due to the lack of a recorded retrieval trajectory. In this paper, we propose a novel iterative RAG method called ReSP, equipped with a dual-function summarizer. This summarizer compresses information from retrieved documents, targeting both the overarching question and the current sub-question concurrently. Experimental results on the multi-hop question-answering datasets HotpotQA and 2WikiMultihopQA demonstrate that our method significantly outperforms the state-of-the-art, and exhibits excellent robustness concerning context length.
Forward citations
Cited by 5 Pith papers
-
From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Retrieval-Augmented Generation Agents Development
A new three-stage pipeline synthesizes executable planner-executor trajectories for RAG agents from QA pairs and evidence, and supervised fine-tuning on these trajectories gives an 8B model large EM gains over publish...
-
BELLE: A Bi-Level Multi-Agent Reasoning Framework for Multi-Hop Question Answering
A question-type-aware, bi-level multi-agent debate that selects and combines existing QA operators outperforms fixed single-method baselines on four multi-hop benchmarks.
-
From Web Search towards Agentic Deep Research: Incentivizing Search with Reasoning Agents
The paper frames agentic deep research as the successor to web search and proposes, without derivation, a test-time scaling law for reasoning and search.
-
Small Encoders Can Rival Large Decoders in Detecting Groundedness
Task-specific encoders (e.g., RoBERTa-large) rival large decoders such as Llama-3-8B and GPT-4o on binary groundedness detection, within 5 to 10 accuracy points while requiring one to three orders of magnitude fewer FLOPs.
-
Continually Self-Improving Language Models for Bariatric Surgery Question--Answering
bRAGgen uses a perplexity threshold to trigger web retrieval and LoRA fine-tuning, improving answers on a new bariatric surgery QA dataset, but the evaluation is confounded by test-time adaptation.
Discussion (0). Continue with ORCID to comment.