# IMHO Fine-Tuning Improves Claim Detection

arXiv: https://arxiv.org/abs/1905.07000 · HTML: https://mlsanity.com/paper/1905.07000/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 May 2019.

## Built on by (most cited first, 27 of 27)

- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [Continual Pre-training of Language Models](https://mlsanity.com/paper/2302.03241/) (2023)
- [Continual Pre-Training of Large Language Models: How to (re)warm your model?](https://mlsanity.com/paper/2308.04014/) (2023)
- [Temporal Adaptation of BERT and Performance on Downstream Document Classification: Insights from Social Media](https://mlsanity.com/paper/2104.08116/) (2021)
- [Efficient Domain Adaptation of Language Models via Adaptive Tokenization](https://mlsanity.com/paper/2109.07460/) (2021)
- [Llama-3.1-FoundationAI-SecurityLLM-Base-8B Technical Report](https://mlsanity.com/paper/2504.21039/) (2025)
- [AFaCTA: Assisting the Annotation of Factual Claim Detection with Reliable LLM Annotators](https://mlsanity.com/paper/2402.11073/) (2024)
- [Medical Adaptation of Large Language and Vision-Language Models: Are We Making Progress?](https://mlsanity.com/paper/2411.04118/) (2024)
- [Adapting a Language Model While Preserving its General Knowledge](https://mlsanity.com/paper/2301.08986/) (2023)
- [LESA: Linguistic Encapsulation and Semantic Amalgamation Based Generalised Claim Detection from Online Content](https://mlsanity.com/paper/2101.11891/) (2021)
- [Downstream Datasets Make Surprisingly Good Pretraining Corpora](https://mlsanity.com/paper/2209.14389/) (2022)
- [Empowering the Fact-checkers! Automatic Identification of Claim Spans on Twitter](https://mlsanity.com/paper/2210.04710/) (2022)
- [ModuleFormer: Modularity Emerges from Mixture-of-Experts](https://mlsanity.com/paper/2306.04640/) (2023)
- [From Chaos to Clarity: Claim Normalization to Empower Fact-Checking](https://mlsanity.com/paper/2310.14338/) (2023)
- [Belief-based Generation of Argumentative Claims](https://mlsanity.com/paper/2101.09765/) (2021)
- [Generating Informative Conclusions for Argumentative Texts](https://mlsanity.com/paper/2106.01064/) (2021)
- [ENTRUST: Argument Reframing with Language Models and Entailment](https://mlsanity.com/paper/2103.06758/) (2021)
- [Lost in Translation, Found in Spans: Identifying Claims in Multilingual Social Media](https://mlsanity.com/paper/2310.18205/) (2023)
- [Utilizing Large Language Models for Information Extraction from Real Estate Transactions](https://mlsanity.com/paper/2404.18043/) (2024)
- [DESYR: Definition and Syntactic Representation Based Claim Detection on the Web](https://mlsanity.com/paper/2108.08759/) (2021)
- [On the Role of Images for Analyzing Claims in Social Media](https://mlsanity.com/paper/2103.09602/) (2021)
- [MM-Claims: A Dataset for Multimodal Claim Detection in Social Media](https://mlsanity.com/paper/2205.01989/) (2022)
- [Overview of the CLAIMSCAN-2023: Uncovering Truth in Social Media through Claim Detection and Identification of Claim Spans](https://mlsanity.com/paper/2310.19267/) (2023)
- [FormLM: Recommending Creation Ideas for Online Forms by Modelling Semantic and Structural Information](https://mlsanity.com/paper/2211.05284/) (2022)
- [Topic Ontologies for Arguments](https://mlsanity.com/paper/2301.09759/) (2023)
- ["Sharks are not the threat humans are": Argument Component Segmentation in School Student Essays](https://mlsanity.com/paper/2103.04518/) (2021)
- [KemenkeuGPT: Leveraging a Large Language Model on Indonesia's Government Financial Data and Regulations to Enhance Decision Making](https://mlsanity.com/paper/2407.21459/) (2024)

## Abstract (arXiv)

Claims are the central component of an argument. Detecting claims across different domains or data sets can often be challenging due to their varying conceptualization. We propose to alleviate this problem by fine tuning a language model using a Reddit corpus of 5.5 million opinionated claims. These claims are self-labeled by their authors using the internet acronyms IMO/IMHO (in my (humble) opinion). Empirical results show that using this approach improves the state of art performance across four benchmark argumentation data sets by an average of 4 absolute F1 points in claim detection. As these data sets include diverse domains such as social media and student essays this improvement demonstrates the robustness of fine-tuning on this novel corpus.