Search papers, labs, and topics across Lattice.
4
0
6
9
Chain-of-Thought prompting, a boon for logical reasoning in LLMs, surprisingly *harms* visual spatial reasoning in multimodal models.
MLLMs still struggle with core visuospatial reasoning skills like abstraction and transformation, lagging far behind human performance on a new cognitive benchmark.
Even when source data is protected, source-free domain adaptation leaks knowledge of source-exclusive classes into the target domain, creating a privacy risk that can be mitigated with adversarial unlearning.
Multimodal models may ace the test, but their reasoning is often a sham: FGRPO makes them explain their answers in a way that actually makes sense.
