SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.08705 · 2026

Analyzing the Correlation Between Hallucinations and Knowledge Conflicts in Large Language Models

Lucrezia Laraspata, Giovanna Castellano, Gennaro Vessio

arXiv · PDF

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 1 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2606.08705.json · badge: /api/ran/2606.08705.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
llaraspata/HallucinationDetection canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
llaraspata/HallucinationDetection/src/model/LlaMa.py
pointer only (licence: NONE) · get_code("30d7eec482ebf6b1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llaraspata/HallucinationDetection when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

apply_rotary_pos_emb Ran
ran; our contract draft was wrong, not the code
llaraspata/HallucinationDetection/src/model/LlaMa.py
pointer only (licence: NONE) · get_code("bac65c3dafaec040")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llaraspata/HallucinationDetection when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_languages Ran
ran; our contract draft was wrong, not the code
llaraspata/HallucinationDetection/src/evaluation/utilis.py
pointer only (licence: NONE) · get_code("09440d7c7275cd7d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llaraspata/HallucinationDetection when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_weight_dir Ran
ran; our contract draft was wrong, not the code
llaraspata/HallucinationDetection/src/model/utils.py
pointer only (licence: NONE) · get_code("338e2303f43212c6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llaraspata/HallucinationDetection when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

rotate_half Ran
ran; our contract draft was wrong, not the code
llaraspata/HallucinationDetection/src/model/LlaMa.py
pointer only (licence: NONE) · get_code("b99eea6376d1e212")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llaraspata/HallucinationDetection when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_tokenizer Not yet run
unverified (harvested, no recorded run)
llaraspata/HallucinationDetection/src/model/utils.py
pointer only (licence: NONE) · get_code("6f394e98c42552fb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llaraspata/HallucinationDetection when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 the same month, 3 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
rotate_half 360 other papers
346 dated earlier, 6 the same month, 5 later, 3 undated · 30 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2302.13971 (2023-02)
LLaMA: Open and Efficient Foundation Language Models
and 357 more not named here
repeat_kv 243 other papers
233 dated earlier, 6 the same month, 2 later, 2 undated · 22 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
and 240 more not named here
apply_rotary_pos_emb 94 other papers
86 dated earlier, 3 the same month, 2 later, 3 undated · 6 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
2309.11674 (2023-09)
A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models
and 91 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Hallucinations, factually incorrect or unverifiable answers, represent one of the most challenging limitations of Large Language Models (LLMs), particularly for knowledge-intensive tasks. According to several studies, one possible explanation lies in the internal knowledge conflicts that arise due to the models' fixed and outdated training data. In this paper, we investigate whether internal representations associated with knowledge conflicts are correlated with hallucination behaviors in LLMs. Using probing techniques inspired by two prior works, we analyzed activations from hidden, attention, and MLP layers, as well as output logits, across predefined tasks to assess the correlation between the two phenomena. Specifically, we probe LLaMA-3-8B on hallucination detection benchmarks and Falcon-7B on a dataset of knowledge conflicts. Our findings indicate that, although these phenomena appear conceptually related, the internal activation patterns of hallucinations cannot be fully reduced to, or explained by, knowledge conflict representations. Nonetheless, we observe that probing remains a robust tool across multiple languages and activation types, supporting its use for improving LLMs' interpretability. This work contributes to the broader understanding of hallucinations in LLMs and highlights the importance of fine-grained analysis of their internal behavior. The source code is publicly available at: https://github.com/llaraspata/HallucinationDetection.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2606.08705")
get_code_for_paper("2606.08705")
have("2606.08705")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2606.08705.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2606.08705.svg)](https://syntology.ai/paper/2606.08705)

Connect an agent — have() is free.