TMLR Editors Quiz Authors Before Desk Rejection; All 10 Suspect Papers Rejected

A novel verification effort by the editorial team of the Transactions on Machine Learning Research (TMLR) sparked wide discussion in the ML community: overwhelmed by submissions and short of reviewers, the journal tightened its desk rejection policy, with the desk rejection rate rising from about 6% in 2023 to about 53%. Co-editor Nihar Shah personally contacted the authors of 10 papers facing desk rejection and quizzed them over Zoom; none of the 7 authors who accepted could answer any question about their own paper, and all 10 papers were ultimately desk-rejected—widely read as a screen for suspected AI-written submissions.

Confirmed

• Facing surging submissions and limited reviewer capacity, TMLR tightened its desk rejection policy; Shah disclosed the rate rose from about 6% in 2023 to roughly 53%.
• Editors contacted authors of 10 papers slated for desk rejection, inviting them to answer questions over Zoom; per @hihey54, one author withdrew the paper outright, one said they had no time, and the remaining 7 accepted.
• None of the 7 who accepted could answer any question about their own paper, as relayed by Delip Rao, @DynamicWebPaige and others; @Pseudomanifold noted the small sample but called the results devastating.
• Per @RexDouglass and @sethlazar, all 10 papers were ultimately desk-rejected, interpreted as the outcome for suspected AI-written submissions.
• Per @TuhinChakr and others, Shah spent about two weeks and 20-25 hours interviewing authors of suspect low-quality submissions; peers including Jessica Hullman praised the effort, with @furongh calling it a hope for saving peer review.
• @pnawrot's thread also noted a researcher's sardonic aside that ablation sections amount to a recipe for p-hacking.

Why it matters

• The approach targets the flood of LLM-era paper mills, ghostwriting, and fraudulent submissions: authors unable to discuss their own papers often did not write or understand them.
• Pre-rejection author quizzes could become a new tool for journals to screen low-quality or suspicious submissions amid reviewer shortages, while raising fairness concerns.
• The jump in desk rejection rate from about 6% to about 53% underscores the severity of the imbalance between submission volume and review capacity in ML publishing.

2026-09-17 ~ 2026-09-19 · 14 related posts

• TMLR tightens desk rejections as submissions surge, quizzes authors on their own papers — thegautamkamath · 2026-09-17
• [source] TMLR's unusual Zoom tests: 7 of 7 invited authors fail to explain their own papers — deliprao · 2026-09-17
• TMLR editor interviewed authors of low-quality submissions: they had no idea what their papers said — TuhinChakr · 2026-09-17
• [source] TMLR desk rejects all 10 suspect papers after author verification calls; 3 couldn't answer basic questions — RexDouglass · 2026-09-17
• TMLR desk rejection rate jumps to 53%, editors quiz authors on their own papers — ducha_aiki · 2026-09-17
• [source] TMLR Quizzed 10 Desk-Rejection Authors; Most Couldn't Explain Their Own Papers — hihey54 · 2026-09-17
• TMLR Zoom defense test: 7 of 7 authors failed to explain their own papers — DynamicWebPaige · 2026-09-17
• TMLR floods with submissions; co-EiC quizzes authors on their own papers — furongh · 2026-09-17
• TMLR Surveyed Authors Before Desk-Rejecting Papers, Results Are Devastating — Pseudomanifold · 2026-09-17
• TMLR tightens desk rejects amid submission deluge, quizzes authors on their own papers — RexDouglass · 2026-09-17
• TMLR desk-rejects all 10 suspected AI-written papers after author calls — sethlazar · 2026-09-17
• TMLR experiment shows spam paper authors don't understand their own work — anmarasovic · 2026-09-17
• TMLR tightens desk rejects amid submission deluge; ablations called a p-hacking recipe — p_nawrot · 2026-09-19

1 near-duplicate retellings: serrjoa