Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2411.16985 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929605626036224 |
|---|---|
| author | Hartill, Tim |
| author_facet | Hartill, Tim |
| contents | Pretrained large Language Models (LLMs) are able to answer questions that are unlikely to have been encountered during training. However a diversity of potential applications exist in the broad domain of reasoning systems and considerations such as latency, cost, available compute resource and internet connectivity are relevant in determining an appropriate approach. We consider the setting where some local compute capacity is available at inference time but internet connectivity is not.
Similar to a general-purpose LLM, we assume that our much smaller Reasoning Models may be asked arbitrary questions from unknown distributions, so we focus on evaluation in an unseen setting. We train our models to answer diverse questions by instilling an ability to reason over a retrieved context. We acquire context from two knowledge sources; a Wikipedia corpus queried using a multi-hop dense retrieval system with novel extensions, and from rationales generated from a larger Language Model optimised to run in a lower resource environment.
Our main contributions: We propose novel methods to show that our model is capable of answering contextualised questions without memorisation. We establish a comprehensive set of baseline results on unseen evaluation datasets. We show that the addition of novel retrieval-augmented training datasets (RATD) to the training regime of the Reasoning Model significantly improves results. We demonstrate further significant improvement through the application of methods for combining knowledge from two sources. The first method (RR) involves training a novel Rationale Ranking model to score both generated rationales and retrieved contexts with respect to relevance and truthfulness. We use the scores to derive combined contexts. We also show that utilising the RATD datasets enables our model to become proficient at utilising combined noisy contexts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_16985 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Teaching Smaller Language Models To Generalise To Unseen Compositional Questions (Full Thesis) Hartill, Tim Computation and Language Artificial Intelligence Pretrained large Language Models (LLMs) are able to answer questions that are unlikely to have been encountered during training. However a diversity of potential applications exist in the broad domain of reasoning systems and considerations such as latency, cost, available compute resource and internet connectivity are relevant in determining an appropriate approach. We consider the setting where some local compute capacity is available at inference time but internet connectivity is not. Similar to a general-purpose LLM, we assume that our much smaller Reasoning Models may be asked arbitrary questions from unknown distributions, so we focus on evaluation in an unseen setting. We train our models to answer diverse questions by instilling an ability to reason over a retrieved context. We acquire context from two knowledge sources; a Wikipedia corpus queried using a multi-hop dense retrieval system with novel extensions, and from rationales generated from a larger Language Model optimised to run in a lower resource environment. Our main contributions: We propose novel methods to show that our model is capable of answering contextualised questions without memorisation. We establish a comprehensive set of baseline results on unseen evaluation datasets. We show that the addition of novel retrieval-augmented training datasets (RATD) to the training regime of the Reasoning Model significantly improves results. We demonstrate further significant improvement through the application of methods for combining knowledge from two sources. The first method (RR) involves training a novel Rationale Ranking model to score both generated rationales and retrieved contexts with respect to relevance and truthfulness. We use the scores to derive combined contexts. We also show that utilising the RATD datasets enables our model to become proficient at utilising combined noisy contexts. |
| title | Teaching Smaller Language Models To Generalise To Unseen Compositional Questions (Full Thesis) |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2411.16985 |