Unverified50% confidenceFactExact time
The study covers five models, two agent frameworks, and three SWE-bench datasets.
1
Sources
50%
Confidence
Long-term
Relevance
11/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study analyzes the Gemma 3 1B and 4B models using Gemma Scope SAEs across eight datasets.75% similarUnverifiedThe study collected multiple tabular datasets of different sizes and different class granularities, and deployed four current state-of-the-art attack methods on multiple model architectures.72% similarUnverifiedThe research team built MIRA-Data to generate high-quality multi-hop multimodal research trajectories.69% similarUnverifiedThe research team constructed two new datasets oriented toward collaborative scenarios and based on memory-grounded QA to evaluate HiCoMER64% similarUnverifiedThe researchers conducted experiments on four real-world datasets covering four different domains; the results show that among the evaluated aggregation methods, SCF achieved the highest AUROC in all four domains.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1165356API
curl https://kongchang.com/api/v1/knowledge/claims/1165356MCP
get_claim(id=1165356)