Unverified50% confidenceBenchmarkExact time
After switching to an eight-token lookup table covering the space-prefixed forms, LLaVA-1.5-7B achieves F1 of 0.8221, 0.8498, and 0.8713 on the adversarial, popular, and random splits respectively; fixing the reading alone moves the adversarial split by 6.13 points.
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 adversarial split F1 measured on LLaVA-1.5-7B is 82.21, 1.21 to 8.67 points higher than the three adversarial baselines compared against (81.00, 79.30, 73.54).77% similarUnverifiedCompared with v1.2, v1.5 improves the F0.5 for real spelling errors from 0.882 to 0.963 and the F0.5 for missing words from 0.078 to 0.698; tense errors remain a weak point, with an F0.5 of only 0.119.66% similarUnverifiedOn a site-disjoint held-out set of 91 cases, SADL achieved a Dice coefficient of 0.7322, a lesion-level F1 of 0.7461, and a PR-AUC of 0.8430.66% similarUnverifiedCriticGen's F1 scores on criteria-based rationales and actionable suggestions improved from 0.6369/0.5994 to 0.7554/0.7900, respectively.65% similarUnverifiedOn the held-out test set, FakeSpotter achieved a macro F1 score of 0.788 for short texts and 0.793 for long texts.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074066API
curl https://kongchang.com/api/v1/knowledge/claims/1074066MCP
get_claim(id=1074066)