Brain-model neural similarity reveals abstractive summarization performance.

Supplementary Table S2 and Table S3 show essentially consistent results for Rouge2 and RougeL, except the positive correlation between RSA scores and Rouge2 decline for the T5 model in Full time window narrowly missed statistical significance ( p = 0.051).

also in 72 other papers

The BART model showed a statistically significant monotonic increase in RSA scores for Full time and Early window ( Z > 0, p < 0.05), but for Late window, the increasing trend did not reach statistical significance ( Z = 1.643, p = 0.100).

also in 34,328 other papers

Conversely, the RS between PEGASUS and T5 hidden layers exhibited a decreasing trend with increasing layer depth.

also in 20,364 other papers

We observed a notable trend: as the model layers deepened, the similarity between encoder layers and human brain language processing generally showed an upward trend, reaching statistical significance ( p < 0.05) under most experimental conditions.

also in 1,536 other papers

The representational similarity between BART and T5 hidden layers showed an increasing trend as the layer depth increased.

also in 25,605 other papers