We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Temperature sampling is a conventional approach to diversify large language model predictions. As temperature increases, the prediction becomes diverse but also vulnerable to hallucinations -- generating tokens that are sensible but not factual. One common approach to mitigate hallucinations is to provide source/grounding documents and the model is trained to produce predictions that bind to and are attributable to the provided source. It appears that there is a trade-off between diversity and attribution. To mitigate any such trade-off, we propose to relax the constraint of having a fixed temperature over decoding steps, and a mechanism to guide the dynamic temperature according to its relevance to the source through KL-divergence. Our experiments justifies the trade-off, and shows that our sampling algorithm outperforms the conventional top-k and top-p algorithms in conversational question-answering and summarization tasks.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2306.01286")
get_code_for_paper("2306.01286")
have("2306.01286")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2306.01286.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2306.01286)
Connect an agent — have() is free.