# SymLoc: Symbolic Localization of Hallucination across HaluEval and TruthfulQA

arXiv: https://arxiv.org/abs/2511.14172 · HTML: https://mlsanity.com/paper/2511.14172/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 18 Nov 2025.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [A Primer in BERTology: What we know about how BERT works](https://mlsanity.com/paper/2002.12327/) (2020)
- [DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models](https://mlsanity.com/paper/2309.03883/) (2023)
- [AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders](https://mlsanity.com/paper/2501.17148/) (2025)
- [Chain-of-Knowledge: Grounding Large Language Models via Dynamic Knowledge Adapting over Heterogeneous Sources](https://mlsanity.com/paper/2305.13269/) (2023)
- [Understanding and Detecting Hallucinations in Neural Machine Translation via Model Introspection](https://mlsanity.com/paper/2301.07779/) (2023)
- [Mechanistic Understanding and Mitigation of Language Model Non-Factual Hallucinations](https://mlsanity.com/paper/2403.18167/) (2024)
- [Towards Robust Evaluation of Unlearning in LLMs via Data Transformations](https://mlsanity.com/paper/2411.15477/) (2024)
- [Detecting and Understanding Vulnerabilities in Language Models via Mechanistic Interpretability](https://mlsanity.com/paper/2407.19842/) (2024)
- [Learning to Paraphrase Sentences to Different Complexity Levels](https://mlsanity.com/paper/2308.02226/) (2023)
- [WellDunn: On the Robustness and Explainability of Language Models and Large Language Models in Identifying Wellness Dimensions](https://mlsanity.com/paper/2406.12058/) (2024)
- [A Domain-Agnostic Neurosymbolic Approach for Big Social Data Analysis: Evaluating Mental Health Sentiment on Social Media during COVID-19](https://mlsanity.com/paper/2411.07163/) (2024)
- [Layer Importance and Hallucination Analysis in Large Language Models via Enhanced Activation Variance-Sparsity](https://mlsanity.com/paper/2411.10069/) (2024)
- [Investigating Symbolic Triggers of Hallucination in Gemma Models Across HaluEval and TruthfulQA](https://mlsanity.com/paper/2509.09715/) (2025)
- [Dense Local Dependencies Induce Attention-Logit Explosion and Training Instability During Long-Sequence Transformer Training](https://mlsanity.com/paper/2505.15548/) (2025)

## Built on by (most cited first, 1 of 1)

- [Hallucinations in LLMs: A Lifecycle-Based Survey of Causes, Detection, Mitigation, and Prevention](https://mlsanity.com/paper/2608.26168/) (2026)

## Abstract (arXiv)

LLMs still struggle with hallucination, especially when confronted with symbolic triggers like modifiers, negation, numbers, exceptions, and named entities. Yet, we lack a clear understanding of where these symbolic hallucinations originate, making it crucial to systematically handle such triggers and localize the emergence of hallucination inside the model. While prior work explored localization using statistical techniques like LSC and activation variance analysis, these methods treat all tokens equally and overlook the role symbolic linguistic knowledge plays in triggering hallucinations. So far, no approach has investigated how symbolic elements specifically drive hallucination failures across model layers, nor has symbolic linguistic knowledge been used as the foundation for a localization framework. We propose the first symbolic localization framework that leverages symbolic linguistic and semantic knowledge to meaningfully trace the development of hallucinations across all model layers. By focusing on how models process symbolic triggers, we analyze five models using HaluEval and TruthfulQA. Our symbolic knowledge approach reveals that attention variance for these linguistic elements explodes to critical instability in early layers (2-4), with negation triggering catastrophic variance levels, demonstrating that symbolic semantic processing breaks down from the very beginning. Through the lens of symbolic linguistic knowledge, despite larger model sizes, hallucination rates remain consistently high (78.3%-83.7% across Gemma variants), with steep attention drops for symbolic semantic triggers throughout deeper layers. Our findings demonstrate that hallucination is fundamentally a symbolic linguistic processing failure, not a general generation problem, revealing that symbolic semantic knowledge provides the key to understanding and localizing hallucination mechanisms in LLMs.