diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-04-16 13:35:31 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-04-16 13:35:31 -0500 |
| commit | 6f48c4fae3243e280b27a977c6a8cb731becf446 (patch) | |
| tree | a0c16715e0634ec819d60df2133627af7677624a /eval | |
| parent | 5c52e69ef80c4081299feba0064530ff8c9eedf5 (diff) | |
Table 1 (main, K=4): 11 baselines + 5 UPH variants (d=8..128) with
state size, inference tokens, R-L±std, METEOR±std on both tasks.
Table 2 (review_k): full K=4/8/16 ROUGE-L for all 15 methods.
Table 3 (topic_k): full K=4/8/16 ROUGE-L for all 15 methods.
d sweep is now folded into each table's UPH block, replacing the
separate small ablation table.
Rewrote §3.2 prose to reflect the flat K-scaling observed universally
on LongLaMP and the low-dimensional nature of the user prior.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Diffstat (limited to 'eval')
0 files changed, 0 insertions, 0 deletions
