Integrating Large Language Models with Graph-based Reasoning for Conversational Question Answering
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916322117419008 |
|---|---|
| author | Jain, Parag Lapata, Mirella |
| author_facet | Jain, Parag Lapata, Mirella |
| contents | We focus on a conversational question answering task which combines the challenges of understanding questions in context and reasoning over evidence gathered from heterogeneous sources like text, knowledge graphs, tables, and infoboxes. Our method utilizes a graph structured representation to aggregate information about a question and its context (i.e., the conversation so far and evidence retrieved to find an answer), while also harnessing the reasoning and text generation capabilities of large language models (LLMs). Graph embeddings are directly injected into the LLM, bypassing the token embedding layers, and learned end-to-end by minimizing cross-entropy. Our model maintains a memory module to track and update past evidence, thus influencing the graph's structure, as the conversation evolves. Experimental results on the ConvMix benchmark(Christmann et al., 2022a) show that graph embeddings enhance the LLM's ability to reason, while the memory module provides robustness against noise and retrieval errors. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_09506 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Integrating Large Language Models with Graph-based Reasoning for Conversational Question Answering Jain, Parag Lapata, Mirella Computation and Language We focus on a conversational question answering task which combines the challenges of understanding questions in context and reasoning over evidence gathered from heterogeneous sources like text, knowledge graphs, tables, and infoboxes. Our method utilizes a graph structured representation to aggregate information about a question and its context (i.e., the conversation so far and evidence retrieved to find an answer), while also harnessing the reasoning and text generation capabilities of large language models (LLMs). Graph embeddings are directly injected into the LLM, bypassing the token embedding layers, and learned end-to-end by minimizing cross-entropy. Our model maintains a memory module to track and update past evidence, thus influencing the graph's structure, as the conversation evolves. Experimental results on the ConvMix benchmark(Christmann et al., 2022a) show that graph embeddings enhance the LLM's ability to reason, while the memory module provides robustness against noise and retrieval errors. |
| title | Integrating Large Language Models with Graph-based Reasoning for Conversational Question Answering |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2407.09506 |