This work presents ingest-time fact compilation, an architecture that performs this work when corpus data is ingested or changed and supports a narrow but practical claim: resolving a corpus state once can make subsequent QA cheaper and more reliable for inexpensive models.
Abstract
Most agentic question answering (QA) systems do an important part of their semantic work at the worst possible time: every time someone asks a question. When a corpus contains revisions, drafts, revocations, deletions, and sources with different levels of authority, the model must reconstruct the governed current state on every read - then throw that work away and repeat it on the next query. This is a bit like a database that rebuilds a materialized view every time someone reads from it. We present ingest-time fact compilation, an architecture that performs this work when corpus data is ingested or changed. Raw passages are rephrased into self-contained facts; rules governing revisions, deletions, effective dates, and source trust are resolved once; and the resulting state is stored as typed records carrying source and revision provenance. At query time, an inexpensive model reads the compiled record instead of reconstructing it from noisy candidates. In a controlled synthetic experiment across five seeds, the same low-cost model produced the correct value, source, and revision in only one of 30 trials under query-time reconstruction, but in all 30 trials from the compiled substrate, at 12.89 times lower mean read cost per question. On simpler revision questions both architectures were exact, but the compiled path used 21.6 times fewer tokens. A separate test found that fact rephrasing roughly halved verbose Federal Reserve dialogue while preserving high source entailment, but left concise Wikipedia prose essentially unchanged. These results support a narrow but practical claim: resolving a corpus state once can make subsequent QA cheaper and more reliable for inexpensive models. We release the open source, MIT-licensed implementation and experimental artifacts.
The paradigm ingest-time semantic compilation (ISC) is called: compile a corpus's meaning into a queryable substrate with two coupled layers - incrementally maintained embeddings, and atomic claims whose provenance is validated at compile time - and treat that substrate as a first-class database object with its own DDL...
Correctly answering a question grounded in normative documents often depends on information outside any single passage: whether the retrieved document is the version currently in force; whether it applies to the jurisdiction, subject (such as an institution or applicant), and date at issue; and whether each normative c...
Liuyin Wang, Shuai-Peng Jin, Ji-Wei Shi et al.· 0 citations
It is argued and shown empirically that maintenance cost scales with the amount of change, not corpus size, and that maintaining, rather than repeatedly reconstructing, a semantic substrate is supported.
Yusuke Takahashi, Kyle Wild, Asako Uraki· Asia Pacific Symposium on In...· 2 citations
MQuAKE-MemStream, a derived dataset of ordered memory streams built from MQuAKE-Remastered counterfactual cases, is introduced, and the Contract Memory Compiler (CMC) is introduced, which achieves state-of-the-art multi-hop accuracy on FactConsolidation.
Zhi Song, Ximing Xing, Chun-Han Li et al.· 2 citations
TIDE, an expert-verified benchmark of 3,050 QA pairs over 644 official customs instruments issued between 1969 and 2025 by the Government of Bangladesh, is presented, covering eight task types over deeply code-mixed documents that are heterogeneous in layout and dated in two calendars.
M. Sobhani, Md. Faiyaz Abdullah Sayeedi, F. Chowdhury et al.· 1 citation· ⚡1
Results show that SQL verification can be performed with a lightweight learned model while retaining feature-level evidence for inspecting and diagnosing its predictions, and feature attribution shows that the model relies on both semantic grounding and deterministic SQL-structure signals.
N. Shukla, Debasmita Panda, Srutanik Bhaduri et al.· 0 citations