ML News
new
|
past
|
best
|
rss
|
submit
about
Stories from July 1, 2026 (UTC)
Go back a
day
,
month
, or
year
. Go forward a
day
.
1.
Reap: Automatic Curation of Coding Agent Benchmarks
(
arxiv.org
)
2 points
by
dipankarsarkar
139 days ago
|
hide
|
past
|
pdf
|
discuss
2.
Discretizing Reward Models
(
arxiv.org
)
2 points
by
gmays
140 days ago
|
hide
|
past
|
pdf
|
discuss
3.
Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems
(
arxiv.org
)
1 point
by
yubblegum
139 days ago
|
hide
|
past
|
pdf
|
discuss
4.
VeriCache: Turning Lossy KV Cache into Lossless LLM Inference
(
arxiv.org
)
1 point
by
matt_d
139 days ago
|
hide
|
past
|
pdf
|
discuss
5.
OpenAI Gym (2016)
(
arxiv.org
)
1 point
by
gregsadetsky
139 days ago
|
hide
|
past
|
pdf
|
discuss
6.
Explaining Attention with Program Synthesis
(
arxiv.org
)
1 point
by
bilsbie
139 days ago
|
hide
|
past
|
pdf
|
discuss
7.
Predictable GRPO
(
arxiv.org
)
1 point
by
rghosh8
139 days ago
|
hide
|
past
|
pdf
|
discuss
8.
Why averaging LLM benchmark scores is fundamentally broken
(
arxiv.org
)
1 point
by
testofschool
139 days ago
|
hide
|
past
|
pdf
|
discuss
9.
Reinforcement Learning with Metacognitive Feedback
(
arxiv.org
)
1 point
by
guard0g
140 days ago
|
hide
|
past
|
pdf
|
1 comment
About
|
RSS
|
RSS (all)
|
HN arXiv