HK
Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Top
New
Best
Ask
Show
Jobs
Request
451.
▲
Adversarial poetry as a universal single-turn jailbreak mechanism in LLMs
arxiv.org
189 comments
7 months ago
capgre
384 points
452.
▲
Phase behavior of Cacio and Pepe sauce
arxiv.org
192 comments
a year ago
rev13013
381 points
453.
▲
A Differentiable Programming System to Bridge ML and Scientific Computing
arxiv.org
71 comments
7 years ago
ChrisRackauckas
381 points
454.
▲
A Brief Introduction to Machine Learning for Engineers (2017)
arxiv.org
38 comments
8 years ago
lainon
379 points
455.
▲
Fair coins tend to land on the same side they started
arxiv.org
265 comments
3 years ago
fbartos
377 points
456.
▲
Simple tasks showing reasoning breakdown in state-of-the-art LLMs
arxiv.org
380 comments
2 years ago
tosh
375 points
457.
▲
LLMs get lost in multi-turn conversation
arxiv.org
259 comments
a year ago
simonpure
374 points
458.
▲
Chain-of-thought can hurt performance on tasks where thinking makes humans worse
arxiv.org
250 comments
2 years ago
benocodes
371 points
459.
▲
Pen and paper exercises in machine learning (2021)
arxiv.org
55 comments
4 years ago
beefman
371 points
460.
▲
Ladder: Self-improving LLMs through recursive problem decomposition
arxiv.org
110 comments
a year ago
fofoz
370 points
461.
▲
Bringing GNU Emacs to Native Code (2020)
arxiv.org
130 comments
5 years ago
textread
369 points
462.
▲
The physics of baking good pizza
arxiv.org
154 comments
8 years ago
Tomte
368 points
463.
▲
There Will Be a Scientific Theory of Deep Learning
arxiv.org
167 comments
2 months ago
jamie-simon
367 points
464.
▲
Player of Games
arxiv.org
231 comments
5 years ago
vatueil
364 points
465.
▲
SkillsBench: Benchmarking how well agent skills work across diverse tasks
arxiv.org
171 comments
4 months ago
mustaphah
364 points
466.
▲
An Empirical Study and Evaluation of Modern CAPTCHAs
arxiv.org
329 comments
3 years ago
vincent_s
362 points
467.
▲
Mixtral 8x7B: A sparse Mixture of Experts language model
arxiv.org
150 comments
2 years ago
ignoramous
359 points
468.
▲
Training language models to be warm and empathetic makes them less reliable
arxiv.org
375 comments
10 months ago
Cynddl
358 points
469.
▲
RWKV: Reinventing RNNs for the Transformer Era
arxiv.org
171 comments
3 years ago
ianbutler
358 points
470.
▲
The universal weight subspace hypothesis
arxiv.org
132 comments
7 months ago
lukeplato
358 points
471.
▲
Smart Algorithms beat Hardware Acceleration for Large-Scale Deep Learning
arxiv.org
135 comments
6 years ago
rkwasny
355 points
472.
▲
Probabilistic Artificial Intelligence
arxiv.org
97 comments
a year ago
pavanto
352 points
473.
▲
Mass Extinction and the Structure of the Milky Way (2013)
arxiv.org
204 comments
9 years ago
DrScump
351 points
474.
▲
Some critical issues with the SWE-bench dataset
arxiv.org
116 comments
a year ago
joshwa
350 points
475.
▲
Deep Forest: Towards an Alternative to Deep Neural Networks
arxiv.org
92 comments
9 years ago
kercker
350 points
476.
▲
Tracking Sex: Implications of widespread data leakage and tracking on porn sites
arxiv.org
222 comments
7 years ago
interweb
348 points
477.
▲
Base64 encoding and decoding at almost the speed of a memory copy
arxiv.org
115 comments
7 years ago
beefhash
344 points
478.
▲
Paxos vs. Raft: Have we reached consensus on distributed consensus?
arxiv.org
65 comments
6 years ago
rbanffy
343 points
479.
▲
Are Open-Source Large Language Models Catching Up?
arxiv.org
212 comments
3 years ago
rkwz
342 points
480.
▲
Mission to reach and operate at the focal region of the solar gravitational lens
arxiv.org
141 comments
4 years ago
WithinReason
340 points
More