Unverified50% confidenceBenchmarkExact time
When using a general-purpose reward model selector, LP-SMC achieves end-to-end accuracy improvements on eleven of twelve settings.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/20/2026
First Seen
Valid until: 1/18/2027
Sources
Related Entities
Related Claims
UnverifiedThe hosted model Jev is significantly more accurate on 9 out of 11 decision points, with margins ranging from +10.8 to +46.0 percentage points.68% similarUnverifiedOn 12 multiple-choice question datasets, LatComm achieves an average accuracy up to 10.2 percentage points higher than the single-vector baseline.66% similarUnverifiedIn the eight-player configuration, the win-rate gap between gpt-oss and the trained network had a one-sided Fisher exact test p = 0.0017, far beyond random fluctuation.66% similarUnverifiedIn the configuration using ten policy snapshots, using only 8 prompts achieved an average accuracy of 24.09%, approaching the 24.51% achieved with 14,080 prompts.66% similarUnverifiedThe paper reports that hybrid reasoning agents achieve accuracy about 10 percentage points higher than conventional models on multiple test sets.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074021API
curl https://kongchang.com/api/v1/knowledge/claims/1074021MCP
get_claim(id=1074021)