# Does the Selected Object Reach the Reader? Auditing Identity Handoffs in Grounded Language-Model Pipelines

arXiv: https://arxiv.org/abs/2609.04579 · HTML: https://mlsanity.com/paper/2609.04579/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 35 references that are in the map.
- First posted on arXiv 4 Sep 2026.

## Builds on (most cited first, 18 of 18)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [ERASER: A Benchmark to Evaluate Rationalized NLP Models](https://mlsanity.com/paper/1911.03429/) (2019)
- [Precise Zero-Shot Dense Retrieval without Relevance Labels](https://mlsanity.com/paper/2212.10496/) (2022)
- [ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction](https://mlsanity.com/paper/2112.01488/) (2021)
- [Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy](https://mlsanity.com/paper/2305.15294/) (2023)
- [HybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Data](https://mlsanity.com/paper/2004.07347/) (2020)
- [PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text](https://mlsanity.com/paper/1904.09537/) (2019)
- [Open Question Answering over Tables and Text](https://mlsanity.com/paper/2010.10439/) (2020)
- [Are self-explanations from Large Language Models faithful?](https://mlsanity.com/paper/2401.07927/) (2024)
- [Efficient One-Pass End-to-End Entity Linking for Questions](https://mlsanity.com/paper/2010.02413/) (2020)
- [Knowledge Guided Text Retrieval and Reading for Open Domain Question Answering](https://mlsanity.com/paper/1911.03868/) (2019)
- [RaFe: Ranking Feedback Improves Query Rewriting for RAG](https://mlsanity.com/paper/2405.14431/) (2024)
- [Open Domain Question Answering with A Unified Knowledge Interface](https://mlsanity.com/paper/2110.08417/) (2021)
- [Multi-Row, Multi-Span Distant Supervision For Table+Text Question](https://mlsanity.com/paper/2112.07337/) (2021)
- [Entity Retrieval for Answering Entity-Centric Questions](https://mlsanity.com/paper/2408.02795/) (2024)
- [AgenticRAGTracer: A Hop-Aware Benchmark for Diagnosing Multi-Step Retrieval Reasoning in Agentic RAG](https://mlsanity.com/paper/2602.19127/) (2026)

## Abstract (arXiv)

Grounded language-model pipelines can be divided into three stages: selecting an object, retrieving passages for it, and using that evidence to answer. If the selected object must reach the reader, losing it breaks the handoff. Benchmark recall checks the dataset-linked object, which can differ. We audit 600 HybridQA questions across three selector families. On 1,463 resolvable records where the selected object matches the dataset-traced passage, exact key lookup and exact title matching return the object every time. With every ranked rule given the same decoded selected title, body-only BM25 omits it on 389 records (26.6%) at cutoff five, while hybrid retrieval with reranking omits it on 14 (1.0%). The two identities differ on 329 of 1,792 resolvable records. With original-question rankings, their top-five checks disagree on 106 records (5.9%). Frozen reader comparisons associate the aligned object's presence with 28.6 to 31.0 points higher exact match. In a deliberately selected 64-item cohort, removing that passage sharply lowers exact match, while removing a similar-length comparison passage does not reproduce the drop. We release the Returned-Object Profile (ROP), an executable record of the target, returned-ID field, cutoff, membership rule, and complete expected population, with data and an offline replay.