Unverified50% confidenceFactExact time
The researchers evaluated 50 MLLMs, and the results show that open-source models as a whole still lag behind closed-source models, and this gap is particularly pronounced on audio-visual understanding tasks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedEvaluation results on current frontier MLLMs show that the models perform strongly on local perception tasks but are clearly weak in long-horizon audio-visual reasoning.74% similarUnverifiedTop academic deepfake detectors can see their accuracy plummet from 99% to below 60% when faced with previously unseen generative models, a phenomenon known as the cross-model generalization problem.64% similarUnverifiedThe gap between selection and generation exceeds 50 percentage points, revealing that reliability is far lower than benchmarks suggest when users need the model to directly write out kinship terms.63% similarUnverifiedThe study finds that across the device–edge–cloud model capability ladder, models at every tier (including the most advanced cloud models) achieve significantly lower accuracy when answering non-standard register queries.63% similarUnverifiedIn the WOVEN evaluation, even the strongest MLLM falls far below human level, and failure modes recur across different model families, so simply scaling up model size cannot eliminate these deficiencies.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1028085API
curl https://kongchang.com/api/v1/knowledge/claims/1028085MCP
get_claim(id=1028085)