CV Researcher Asks: With Thousands of Papers a Year, How Much Novelty Is Left?
ATHii-127 · reddit · 2026-10-01
A computer vision researcher reports that 'novelty' concerns recur across years of submissions to NeurIPS, ICLR and CVPR. Given thousands of top-conference papers annually, they question how much genuinely new territory remains and seek advice on framing contributions, separating meaningful incremental progress from insufficiently novel work, and what reviewers actually look for.
More from Research
- Researcher teases 'interp exponential': wave of interpretability papers coming in weeks — burny_tech · 2026-10-01
- DerivAudit: 17-21% of long-term agent memories unsupported by interaction history — newyorkuniversity · 2026-10-01
- VoxParity benchmark: only 11 of 23 voice agents act on what they hear, not just read — Bhavik Mangla · 2026-10-01
- CASCADE filters genetic variants through four questions; the 3% that pass are twice as GWAS-enriched — anshulkundaje · 2026-10-01
- Nature paper: 10M-cell immune multiome atlas from 1,108 FinnGen donors links disease variants to regulation — anshulkundaje · 2026-10-01
- Duplicate repost of the same Nature immune multiome atlas paper — anshulkundaje · 2026-10-01