# CARAT: Do Materials LLMs Reason or Recite?

arXiv: https://arxiv.org/abs/2609.38340 · HTML: https://mlsanity.com/paper/2609.38340/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 29 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 15 of 15)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Faith and Fate: Limits of Transformers on Compositionality](https://mlsanity.com/paper/2305.18654/) (2023)
- [MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields](https://mlsanity.com/paper/2206.07697/) (2022)
- [Causal Abstractions of Neural Networks](https://mlsanity.com/paper/2106.02997/) (2021)
- [State of What Art? A Call for Multi-Prompt LLM Evaluation](https://mlsanity.com/paper/2401.00595/) (2023)
- [Fine-Tuned Language Models Generate Stable Inorganic Materials as Text](https://mlsanity.com/paper/2402.04379/) (2024)
- [Large Language Models are Good Relational Learners](https://mlsanity.com/paper/2506.05725/) (2025)
- [LLM4Mat-Bench: Benchmarking Large Language Models for Materials Property Prediction](https://mlsanity.com/paper/2411.00177/) (2024)

## Abstract (arXiv)

When a materials LLM answers a question about crystal structure, does it reason from the structure or copy an answer already printed in its input? Accuracy cannot tell: a structural description often prints the very field it is scored against. CARAT holds question and gold answer fixed across eight matched views, names each structural relation separately in GraphSpace, and adds matched fine-tuning, answer masking, evidence injection, paired inference, and a rule that can withhold claims. First, on the benchmark's hardest families the grounded view is worth 17.3 points over formula inputs. Second, we turn that scrutiny on ourselves. GraphSpace beats a plain periodic graph by 19.3 points, but that margin is two effects at once: where the plain rendering carries everything the question needs it is 1.96 points, and where it omits those fields entirely, 46.7 points. The headline mostly measures what the baseline lacked, not how evidence is presented. Third, we attack our own benchmark. A rule that skips the link and reads the list directly answers four of seven hardened families, so we rebuilt it until eleven such shortcuts sat near chance. The frozen model quotes that link yet answers the same when we redirect it, on 95.6% of paired cases: it repeats the relation without using it. After matched supervision it reaches 99.8%, and deleting the link drops it to 23.4%, below the 27.0% the best shortcut reaches: both steps are learnable.