TensorX

Trends · 研究趋势

研究趋势

Hugging FaceTensorX精选

数据来自 Hugging Face 论文的 AI 提取关键词,按月统计研究方向的增长与热度。

WeekMonthYear趋势
返回趋势

on-policy distillation 相关论文

23 篇论文 · 按点赞排序

21

Miles v0.1: Production-Level Post-Training

RadixArk, Tom Chen, Mao Cheng +10 authors

Miles is an open-source, production-ready system for large-scale reinforcement learning and post-training that supports diverse backends, weight synchronization, LoRA, distillation, and diffusion models.

57reinforcement-learningrollout enginesHF ↗arXiv ↗
23

Cliff: Learning Process Rewards from the First Mistake

Peixuan Han, Runhui Wang, Ketan Ramaneti +3 authors

Cliff improves reinforcement learning with verifiable rewards by using an off-the-shelf language model to detect the first reasoning error and shaping token-level advantages accordingly.

15reinforcement learning with verifiable rewardsprocess reward modelingHF ↗arXiv ↗
2 / 2 页

上升最快

近 6 个月
1
large language models (llms)NEW
56 篇论文
2
in-context learningNEW
38 篇论文
3
mllmsNEW
29 篇论文
4
mathematical reasoningNEW
26 篇论文
5
multimodal large language modelNEW
17 篇论文
6
diffusion transformerNEW
15 篇论文
7
distillationNEW
14 篇论文
8
ditNEW
13 篇论文
9
zero-shotNEW
12 篇论文
10
curriculum learningNEW
12 篇论文
11
next-token predictionNEW
12 篇论文
12
cross-attentionNEW
11 篇论文
13
multi-agent systemsNEW
11 篇论文
14
contrastive learningNEW
10 篇论文
15
reward hackingNEW
10 篇论文
16
llm-as-a-judgeNEW
10 篇论文
17
multimodal benchmarksNEW
10 篇论文
18
chain-of-thought+500%
30 篇论文
19
rlvr+400%
15 篇论文
20
recursive self-improvement+400%
12 篇论文

最热方向

按总量
1
large language models
322 篇论文
2
reinforcement learning
221 篇论文
3
llms
167 篇论文
4
supervised fine-tuning
94 篇论文
5
diffusion models
76 篇论文
6
multimodal large language models
75 篇论文
7
fine-tuning
75 篇论文
8
vision-language models
61 篇论文
9
large language models (llms)
56 篇论文
10
gpt-4
49 篇论文
11
grpo
44 篇论文
12
mixture-of-experts
40 篇论文
13
in-context learning
38 篇论文
14
large language model
38 篇论文
15
llm
35 篇论文
16
reasoning
35 篇论文
17
language models
33 篇论文
18
vision-language model
31 篇论文
19
chain-of-thought
30 篇论文
20
video generation
29 篇论文
21
mllms
29 篇论文
22
lora
29 篇论文
23
multimodal models
29 篇论文
24
parameter-efficient fine-tuning
28 篇论文
25
benchmark
27 篇论文
26
gpt-4o
27 篇论文
27
image generation
26 篇论文
28
mathematical reasoning
26 篇论文
29
text-to-image generation
25 篇论文
30
instruction tuning
24 篇论文
31
benchmarks
23 篇论文
32
on-policy distillation
23 篇论文
33
sft
23 篇论文
34
transformers
23 篇论文
35
pre-training
23 篇论文
36
transformer
22 篇论文
37
latent space
20 篇论文
38
code generation
20 篇论文
39
llama
20 篇论文
40
vlms
20 篇论文
41
gsm8k
20 篇论文
42
vision-language-action models
19 篇论文
43
image editing
19 篇论文
44
hallucinations
19 篇论文
45
diffusion model
19 篇论文
46
group relative policy optimization
18 篇论文
47
rl
18 篇论文
48
reinforcement learning with verifiable rewards
18 篇论文
49
retrieval-augmented generation
18 篇论文
50
multimodal reasoning
18 篇论文
51
multimodal large language model
17 篇论文
52
low-rank adaptation
17 篇论文
53
temporal consistency
17 篇论文
54
speculative decoding
16 篇论文
55
post-training
16 篇论文
56
multimodal understanding
16 篇论文
57
text-to-image
16 篇论文
58
stable diffusion
16 篇论文
59
world models
16 篇论文
60
computational efficiency
16 篇论文

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号