Small Encoders Can Rival Large Decoders in Detecting Groundedness
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908423700873216 |
|---|---|
| author | Abbes, Istabrak Prato, Gabriele Fournier, Quentin Rodriguez, Fernando Boukhary, Alaa Elwood, Adam Chandar, Sarath |
| author_facet | Abbes, Istabrak Prato, Gabriele Fournier, Quentin Rodriguez, Fernando Boukhary, Alaa Elwood, Adam Chandar, Sarath |
| contents | Augmenting large language models (LLMs) with external context significantly improves their performance in natural language processing (NLP) tasks. However, LLMs struggle to answer queries reliably when the provided context lacks information, often resorting to ungrounded speculation or internal knowledge. Groundedness - generating responses strictly supported by the context - is essential for ensuring factual consistency and trustworthiness. This study focuses on detecting whether a given query is grounded in a document provided in context before the costly answer generation by LLMs. Such a detection mechanism can significantly reduce both inference time and resource consumption. We show that lightweight, task specific encoder models such as RoBERTa and NomicBERT, fine-tuned on curated datasets, can achieve accuracy comparable to state-of-the-art LLMs, such as Llama3 8B and GPT4o, in groundedness detection while reducing inference latency by orders of magnitude. The code is available at : https://github.com/chandarlab/Hallucinate-less |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_21288 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Small Encoders Can Rival Large Decoders in Detecting Groundedness Abbes, Istabrak Prato, Gabriele Fournier, Quentin Rodriguez, Fernando Boukhary, Alaa Elwood, Adam Chandar, Sarath Computation and Language Artificial Intelligence Information Retrieval Machine Learning Augmenting large language models (LLMs) with external context significantly improves their performance in natural language processing (NLP) tasks. However, LLMs struggle to answer queries reliably when the provided context lacks information, often resorting to ungrounded speculation or internal knowledge. Groundedness - generating responses strictly supported by the context - is essential for ensuring factual consistency and trustworthiness. This study focuses on detecting whether a given query is grounded in a document provided in context before the costly answer generation by LLMs. Such a detection mechanism can significantly reduce both inference time and resource consumption. We show that lightweight, task specific encoder models such as RoBERTa and NomicBERT, fine-tuned on curated datasets, can achieve accuracy comparable to state-of-the-art LLMs, such as Llama3 8B and GPT4o, in groundedness detection while reducing inference latency by orders of magnitude. The code is available at : https://github.com/chandarlab/Hallucinate-less |
| title | Small Encoders Can Rival Large Decoders in Detecting Groundedness |
| topic | Computation and Language Artificial Intelligence Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2506.21288 |