Benjamin Van Durme, Jeffrey Cheng, William Jurayj
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.
Scaling the test-time compute of large language models has demonstrated impressive performance on reasoning benchmarks. However, existing evaluations of test-time scaling make the strong assumption that a reasoning system should always give an answer to any question provided. This overlooks concerns about whether a model is confident in its answer, and whether it is appropriate to always provide a response. To address these concerns, we extract confidence scores during reasoning for thresholding model responses. We find that increasing compute budget at inference time not only helps models answer more questions correctly, but also increases confidence in correct responses. We then extend the current paradigm of zero-risk responses during evaluation by considering settings with non-zero levels of response risk, and suggest a recipe for reporting evaluations under these settings. 1
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2502.13962")
get_code_for_paper("2502.13962")
have("2502.13962")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2502.13962.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2502.13962)
Connect an agent — have() is free.