Hierarchical Sequence Iteration for Heterogeneous Question Answering
This work addresses the problem of improving accuracy and efficiency in question answering across diverse data formats for AI systems, representing a novel method for a known bottleneck rather than an incremental improvement.
The paper tackles the brittleness of retrieval-augmented generation on multi-step questions with heterogeneous evidence sources by introducing Hierarchical Sequence Iteration, a unified framework that linearizes documents, tables, and knowledge graphs into reversible sequences and performs structure-aware iteration to collect evidence before answer synthesis. Experiments on datasets like HotpotQA and HybridQA show consistent EM/F1 gains over strong baselines with high efficiency.
Retrieval-augmented generation (RAG) remains brittle on multi-step questions and heterogeneous evidence sources, trading accuracy against latency and token/tool budgets. This paper introducesHierarchical Sequence (HSEQ) Iteration for Heterogeneous Question Answering, a unified framework that (i) linearize documents, tables, and knowledge graphs into a reversible hierarchical sequence with lightweight structural tags, and (ii) perform structure-aware iteration to collect just-enough evidence before answer synthesis. A Head Agent provides guidance that leads retrieval, while an Iteration Agent selects and expands HSeq via structure-respecting actions (e.g., parent/child hops, table row/column neighbors, KG relations); Finally the head agent composes canonicalized evidence to genearte the final answer, with an optional refinement loop to resolve detected contradictions. Experiments on HotpotQA (text), HybridQA/TAT-QA (table+text), and MetaQA (KG) show consistent EM/F1 gains over strong single-pass, multi-hop, and agentic RAG baselines with high efficiency. Besides, HSEQ exhibits three key advantages: (1) a format-agnostic unification that enables a single policy to operate across text, tables, and KGs without per-dataset specialization; (2) guided, budget-aware iteration that reduces unnecessary hops, tool calls, and tokens while preserving accuracy; and (3) evidence canonicalization for reliable QA, improving answers consistency and auditability.