--- title: "Diagnosable ColBERT — paper by Remy (2026) · BioNLP 2026" description: "This short position paper proposes Diagnosable ColBERT. 2 references, cited by 2 papers here." authors: ["François Remy"] year: 2026 venue: "BioNLP 2026" source: "https://aclanthology.org/2026.bionlp-1.58/" doi: "10.18653/v1/2026.bionlp-1.58" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/remy-2026-diagnosable-colbert-debugging-late-interaction/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/remy-2026-diagnosable-colbert-debugging-late-interaction/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/7/) / Diagnosable ColBERT: Debugging Late-Interaction Retrieval Models Using a Learned Latent Space as Reference # Diagnosable ColBERT: Debugging Late-Interaction Retrieval Models Using a Learned Latent Space as Reference **Authors:** [François Remy](https://md.research.elamur.ai/free/en/authors/francois-remy/) · **Year:** 2026 · **Venue:** [BioNLP 2026](https://md.research.elamur.ai/free/en/venues/bionlp/) ([BioNLP 2026](https://md.research.elamur.ai/free/en/venues/bionlp/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.bionlp-1.58/) · [PDF](https://aclanthology.org/2026.bionlp-1.58.pdf) · [DOI 10.18653/v1/2026.bionlp-1.58](https://doi.org/10.18653/v1/2026.bionlp-1.58) · [OpenAlex](https://openalex.org/W7166873515) **Citations:** 0 citations (OpenAlex, 2026-10-14) · [2](https://md.research.elamur.ai/free/en/papers/remy-2026-diagnosable-colbert-debugging-late-interaction/cited-by/) in our library, of the 149,439 papers whose bibliographies we read ## Summary ColBERT-style late-interaction retrievers are often called interpretable because they show which query token matched which document token. The author argues that this is not enough for clinical search. These scores justify a successful match after the fact, but they say little about why a relevant document was missed. They also cannot show whether the model reliably grasps a concept and its qualifiers (negation, temporality, uncertainty, experiencer) across different wordings and contexts. This short position paper proposes Diagnosable ColBERT. Token embeddings are mapped through pre-projection adapters into a reference latent space shaped by clinical knowledge and expert similarity constraints, such as BioLORD-like spaces. Retrieval vectors are then derived as a lower-dimensional projection of this inspectable representation. Testers can see where query and document tokens fall in the space and tell concept-grounding errors apart from contextual misreadings, sometimes without issuing any query. The paper reports no quantitative evaluation. Its contribution is the diagnostic framing plus practitioner experience. This matters because it turns retrieval misses into targeted data-curation decisions. ## Key points - Published at the BioNLP 2026 workshop (pages 713–717) by François Remy (Parallia Healthcare). It is a position paper with no benchmark numbers. - The architecture couples diagnosis and retrieval: the diagnosed embedding lives in the reference space, and the ranking embedding is a down-projection of it. - Worked example: the query *bartonellosis* misses a report that mentions only *CSD* (cat scratch disease). Checking each side's placement in the latent space shows whether the fault lies in query grounding, document-side abbreviation handling, or both. - Query-free diagnosis is also possible. In a dialogue where a patient reports an allergy to *ranitidine*, the tester can check whether the token encodes the allergy context or only the drug identity. - The author reports that debugging interfaces built on this idea guided the move from ClinicalEncoder25 to ClinicalEncoder26AM. Public demos are at demo26.parallia.eu and text2json.parallia.eu. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* 📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,437 works 🧭 Topics (1): [🔍 Information retrieval & search](https://md.research.elamur.ai/free/en/topics/information-retrieval/) · 473 works ## Cite Cite as: François Remy. 2026. Diagnosable ColBERT: Debugging Late-Interaction Retrieval Models Using a Learned Latent Space as Reference. BioNLP 2026. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{remy2026diagnosable, title = {{Diagnosable ColBERT: Debugging Late-Interaction Retrieval Models Using a Learned Latent Space as Reference}}, author = {François Remy}, booktitle = {BioNLP 2026}, year = {2026}, doi = {10.18653/v1/2026.bionlp-1.58}, url = {https://doi.org/10.18653/v1/2026.bionlp-1.58} } ``` - [← previous Detecting Primary Progressive Aphasia (PPA) from Text: A Benchmarking Study](https://md.research.elamur.ai/free/en/papers/merhbene-2026-detecting-primary-progressive-aphasia-ppa/) - [next → Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMs](https://md.research.elamur.ai/free/en/papers/shmidman-2026-dicta-lm-3-0-advancing/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/7/) ## References 2 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Walter S. Campbell, James R. Campbell, William W. West, James C. McClay, and Steven H. Hinrichs. 2014. Semantic analysis of SNOMED CT for a post-coordinated database of histopathology findings. Journal of the American Medical Informatics Association, 21(5):885–892. [find](https://scholar.google.com/scholar?q=Walter+S%2E+Campbell%2C+James+R%2E+Campbell%2C+William+W%2E+West%2C+James+C%2E+McClay%2C+and+Steven+H%2E+Hinrichs%2E+2014%2E+Semantic+analysis+of+SNOMED+CT+for+a+post%2Dcoordinated+database+of+histopathology+findings%2E+Journa) 2. Jens-Joris Decorte, Jeroen Van Hautte, Chris Develder, and Thomas Demeester. 2025. Efficient text encoders for labor market analysis. IEEE Access, 13:133596– 133608. [find](https://scholar.google.com/scholar?q=Jens%2DJoris+Decorte%2C+Jeroen+Van+Hautte%2C+Chris+Develder%2C+and+Thomas+Demeester%2E+2025%2E+Efficient+text+encoders+for+labor+market+analysis%2E+IEEE+Access%2C+13%3A133596%E2%80%93+133608%2E) *Updated 1 November 2026*