Vivek Hebbar

• Vivek Hebbar 8 Aug 2026 9:46 UTC

  3 points

  −16

  in reply to: Vivek Hebbar’s comment on: Ar­gu­ments for P

  What does the disagree vote even mean? 😆

• Vivek Hebbar 8 Aug 2026 1:48 UTC

  12 points

  −7

  on: Ar­gu­ments for P

  Shakespeare: To P or not to P, that is the question

• Vivek Hebbar 7 Aug 2026 22:17 UTC

  2 points

  0

  in reply to: Vivek Hebbar’s comment on: In­di­vi­d­u­ally in­cen­tivized safe Pareto im­prove­ments in open-source bargaining

  Also: is the text in this section meant to express a coherent argument on its own or just be a lossy summary of the more detailed argument in later sections? If just the latter then maybe it’s not worth answering my parent question.

• Vivek Hebbar 7 Aug 2026 22:02 UTC

  2 points

  0

  on: In­di­vi­d­u­ally in­cen­tivized safe Pareto im­prove­ments in open-source bargaining

  Each player independently commits to a bargaining strategy that will be played regardless of whether the bargaining game is or .

  • What exactly do we mean by a “bargaining strategy” here? What restrictions does it place on the ways in which the player’s behavior can depend on the game?

  • Are we assuming that this bargaining strategy behaves identically when we change the disagreement point from to ? If so, why would that be a valid assumption?

Ad­vice for mak­ing ro­bust-to-train­ing model organisms

Why does off-model SFT de­grade ca­pa­bil­ities?

In­crim­i­nat­ing mis­al­igned AI mod­els via distillation

Re­search Sab­o­tage in ML Codebases

Sleeper Agent Back­door Re­sults Are Messy

An Em­piri­cal Study of Meth­ods for SFTing Opaque Rea­son­ing Models

How do LLMs gen­er­al­ize when we do train­ing that is in­tu­itively com­pat­i­ble with two off-dis­tri­bu­tion be­hav­iors?

Five ap­proaches to eval­u­at­ing train­ing-based con­trol measures

Model or­ganisms re­searchers should check whether high LRs defeat their model organisms

• Vivek Hebbar 14 Mar 2026 5:01 UTC

  LW: 2 AF: 1

  0

  AF

  in reply to: Lukas Finnveden’s comment on: Oper­a­tional­iz­ing FDT

  I did say “suppose you are deterministic”. That said, can you spell out how CDT ratifies the optimal policy if randomization is allowed?

• Vivek Hebbar 14 Mar 2026 5:00 UTC

  8 points

  3

  in reply to: cubefox’s comment on: Oper­a­tional­iz­ing FDT

  • It’s impossible for having a proof of A to make it harder to prove B.

  • Proof(A&B) should not be longer than Proof(A) and Proof(B) put together, since proving A and B separately also proves A&B. (Except maybe you waste a few tokens of syntax to say “therefore A&B” or something.)

Oper­a­tional­iz­ing FDT

• Vivek Hebbar 25 Feb 2026 10:16 UTC

  2 points

  0

  in reply to: Cleo Nardo’s comment on: A sim­ple rule for causation

  Thanks, good idea!

A sim­ple rule for causation

How will we do SFT on mod­els with opaque rea­son­ing?

The­o­ret­i­cal pre­dic­tions on the sam­ple effi­ciency of train­ing poli­cies and ac­ti­va­tion monitors