arXiv:2608. 08512v1 Announce Type: new Abstract: Evolving documents, such as laws, tax codes, and software documentation, are amended, replaced, and sometimes reverted over time, so a question has different correct answers at different dates.
The paper introduces "stale‑document poisoning," a temporal alignment failure where outdated retrieval evidence causes retrieval‑augmented generation models to produce incorrect answers even when the model could answer correctly without retrieval. A benchmark of 317 verified knowledge reversals across medicine, law, software, and platform policy shows that outdated evidence flips 30–37% of Llama and Qwen answers, rising to 66–75% when models are explicitly instructed to trust the document. The study demonstrates that providing explicit validity information and a recency‑aware re‑ranker can substantially reduce poisoning, highlighting the need for models to assess whether retrieved evidence remains applicable.
The paper introduces ingest‑time fact compilation, an architecture that preprocesses and compiles corpus data into self‑contained facts with resolved revisions, deletions, and source trust. By storing this compiled state, query‑time models can retrieve answers directly, avoiding costly reconstruction from raw passages. Experiments show that this approach reduces read cost per question by 12.89× and token usage by 21.6× while maintaining accuracy.
arXiv:2607. 16019v1 Announce Type: new Abstract: AI systems increasingly retrieve from records that revise themselves: issue threads, encyclopedic histories, policy logs, and long conversations.
The study shows that in memory‑agent benchmarks, simply retrieving the correct fact does not guarantee accurate answers. On the FactConsolidation task, increasing the evidence budget from 64 to 512 tokens caused accuracy to drop, even though the newest statement was present in every retrieved set. Experiments with Qwen models revealed that replacing an unrelated statement with an obsolete competing statement drastically reduced accuracy, indicating that outdated information can be more harmful than irrelevant evidence.
The paper introduces Knowledge Pull Requests (KPRs), a framework that enables continual document authoring by making each change interpretable. KPRs extract claims from new knowledge sources, filter and route them to appropriate sections, and flag conflicts with existing content, producing a ChangeLog that separates knowledge changes from textual edits. Experiments on revising Wikipedia and updating query-driven reports show that KPRs integrate more information, better preserve existing content, and improve question answering performance compared to rewriting from scratch or using frontier models with search.
arXiv:2610.32109v1 Announce Type: new Abstract: Language models can answer a question from a supplied document or from their parametric knowledge. When later asked where the answer came from, what de...
arXiv:2608. 09393v1 Announce Type: cross Abstract: We identify and quantify temporal misgrounding: the systematic retrieval and citation of the currently in-force version of a legal article when the applicable version is an earlier or future one.
arXiv:2610.15882v1 Announce Type: new Abstract: A stored fact may stop being valid when a user changes a preference or an event replaces an earlier one. We study whether a language model can learn th...
The paper introduces Temporal State Tracking Memory (TEM), a training‑free framework that stores facts as versioned records linked to their sources and retrieves the relevant versions in source order, labeling historical records before passing them to the answer model. TEM incorporates event memory and original dialogue turns to fill gaps in structured facts. Evaluations on 420 questions from 600 sessions with 60 personas show high key‑entity hit rates (75.71–97.62%) and significant improvements when historical labels are added, while keeping evidence compact (94–118 tokens per answer).
The paper investigates how entity‑replacement responses can be reused across different English factual questions when model weights are fixed. By injecting hidden‑state changes from two source questions into a target question about the same entity, the authors compare predicted log‑probability shifts with actual measurements, finding that hidden reuse can reduce no‑change error by up to 49% on certain models and that a later protocol can achieve 58‑68% success on three cohorts. The study also notes that larger checkpoints, chat formats, and new entities affect replication, and that relation‑neutral changes at entity tokens lower relation‑selective error more effectively than random changes.
arXiv:2608.22856v1 Announce Type: cross Abstract: A retrieval-augmented QA system can return different answers after an index expansion even when its requested model identifier, prompt, retrieval pol...