99 Papers Reviews Portable Info

Aris’s mouth went dry. He had forgotten about #033. The one where he just scanned the equations.

He never fixed the sink. But the next conference cycle, he accepted only three papers to review. And he read every word.

“Aris,” she said, her voice brittle. “Did you… read these?” 99 papers reviews

Aris had a secret. For the last ten years, he had been training a personal AI—a small, local language model he called “Erasmus.” He fed Erasmus every review he had ever written. Every terse critique. Every cutting remark about “insufficient novelty” or “flawed experimental design.”

But the other 96? Erasmus ate them. Reviews full of sterile, correct, utterly meaningless jargon flooded the submission system. “The state diagram in Figure 4 lacks clarity.” “The baseline comparison in Table 2 is underpowered.” “The authors should consider a sensitivity analysis.” Aris’s mouth went dry

“We’re accepting their paper, obviously,” she said. “But I need to know, Aris. Did you outsource your reviews to an AI?”

Dr. Aris Thorne was a man built of deadlines. For twenty years, he had been a pillar of the computational linguistics community, a full professor at a respected university, and the go-to reviewer for three top-tier journals. His colleagues called him "The Last Cigarette" because he burned slow, steady, and left a lingering, acrid presence on every paper he touched. He never fixed the sink

His wife left a note on the fridge: “You promised to fix the sink.” He ignored it. His graduate students sent panicked emails about their own theses. He archived them.