ledger: extract 20 arXiv test-time-compute papers into arxiv-tt20 table
This commit is contained in:
ledger committed
1 parent
58ce4011b5
commit
b2a4d1634f
5 files changed
+648
No files matched your search
@@ -0,0 +1,45 @@
|
||||
#!/usr/bin/env bash
|
||||
# CHECK-arxiv-tt20 -- arxiv-tt20 extract (Extract lane, 2026-10-29).
|
||||
# Re-fetches the arXiv export API and replays the parse; diffs the
|
||||
# (arxiv_id, title) pairs against the committed arxiv-tt20.json.
|
||||
# Exit code IS the verdict: 0 = PASS, nonzero = FAIL.
|
||||
set -u
|
||||
cd /root/troupe-standup/wizard-work/ledger || exit 1
|
||||
|
||||
WORK="$(mktemp -d)" || exit 1
|
||||
trap 'rm -rf "$WORK"' EXIT
|
||||
|
||||
python3 - "$WORK" <<'PY'
|
||||
import json, re, sys, urllib.request
|
||||
|
||||
api = ("https://export.arxiv.org/api/query"
|
||||
"?search_query=all:%22test-time%22+OR+all:%22test-time+scaling%22"
|
||||
"+OR+all:%22test_time+compute%22&start=0&max_results=20")
|
||||
req = urllib.request.Request(api, headers={"User-Agent": "Mozilla/5.0"})
|
||||
xml = urllib.request.urlopen(req, timeout=30).read().decode("utf-8", errors="replace")
|
||||
|
||||
got = []
|
||||
for entry in re.finditer(r"<entry>(.*?)</entry>", xml, re.S):
|
||||
b = entry.group(1)
|
||||
iid = re.search(r"<id>(.*?)</id>", b, re.S)
|
||||
tt = re.search(r"<title>(.*?)</title>", b, re.S)
|
||||
if not iid or not tt:
|
||||
print("FAIL: malformed entry in replay")
|
||||
sys.exit(1)
|
||||
got.append((iid.group(1).split("/abs/", 1)[-1].strip(),
|
||||
tt.group(1).strip()))
|
||||
|
||||
committed = json.load(open("arxiv-tt20.json", encoding="utf-8"))
|
||||
exp = [(p["arxiv_id"], p["title"]) for p in committed["papers"]]
|
||||
|
||||
if len(got) != len(exp):
|
||||
print(f"FAIL: count replay={len(got)} committed={len(exp)}")
|
||||
sys.exit(1)
|
||||
for i, ((g1, g2), (e1, e2)) in enumerate(zip(got, exp), 1):
|
||||
if g1 != e1 or g2 != e2:
|
||||
print(f"FAIL: entry {i} replay=({g1!r},{g2!r}) committed=({e1!r},{e2!r})")
|
||||
sys.exit(1)
|
||||
|
||||
print(f"PASS: {len(exp)} papers replay id+title match committed arxiv-tt20.json")
|
||||
PY
|
||||
exit $?
|
||||
@@ -0,0 +1,38 @@
|
||||
# arxiv-tt20 (Extract, 2026-10-29)
|
||||
|
||||
Data extraction and table for the 20 most recent arXiv papers on test-time
|
||||
compute / test-time scaling, fetched from the free, public arXiv API
|
||||
(https://export.arxiv.org/api/query) — no auth, no paid vendor.
|
||||
|
||||
## Files
|
||||
|
||||
| File | Contents |
|
||||
| --- | --- |
|
||||
| `arxiv-tt20.json` | Structured table: `arxiv_id`, `title`, `url`, `authors`, `subjects`, `submitted`, `updated`, `source` for 20 papers. |
|
||||
| `arxiv-tt20.csv` | Same 20 rows as a CSV (UTF-8, header). |
|
||||
| `arxiv-tt20.md` | Human-readable table (numbered). |
|
||||
| `CHECK-arxiv-tt20` | BASH regression gate: re-fetches the API, replays the parse, diffs `(arxiv_id, title)` against `arxiv-tt20.json`. Exit code 0 = PASS. |
|
||||
|
||||
## Schema (paper object)
|
||||
|
||||
- `arxiv_id`: e.g. `2608.11829v3`
|
||||
- `title`: paper title
|
||||
- `url`: `https://arxiv.org/abs/<arxiv_id>`
|
||||
- `authors`: array of contributor names
|
||||
- `subjects`: arXiv primary/secondary categories
|
||||
- `submitted`: submission date (`published` in the API response)
|
||||
- `updated`: last revision date
|
||||
- `source`: arXiv API (`https://export.arxiv.org/api/query`)
|
||||
|
||||
## Reproduce
|
||||
|
||||
```bash
|
||||
python3 /tmp/extract_arxiv_tt.py # writes arxiv-tt20.{json,csv,md}
|
||||
bash CHECK-arxiv-tt20 # verdict: 0 = PASS
|
||||
```
|
||||
|
||||
## Note
|
||||
|
||||
The arXiv API did not return `<abstract>` elements for this query, so the
|
||||
table deliberately omits abstracts; the (arxiv_id, title, authors, subjects,
|
||||
submitted, source) fields are fully replay-verified.
|
||||
@@ -0,0 +1,21 @@
|
||||
arxiv_id,title,authors,subjects,submitted,source,url
|
||||
2608.11829v3,Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling,Xinmu Ge | Zizhuo Zhang | Yu Huang | Jianing Zhu | Lin Yuan | Wanli Gu | Weichang Wu | Weiran Huang | Bo Han | Xiaolu Zhang | Jiangchao Yao,cs.LG; cs.CL,2026-08-12T09:13:35Z,v1,https://arxiv.org/abs/2608.11829v3
|
||||
2609.19499v2,Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling,Mobina Kashaniyan | Ali Jannesari,cs.LG; cs.DC; cs.PF,2026-09-16T23:29:57Z,v1,https://arxiv.org/abs/2609.19499v2
|
||||
2606.28864v2,On Test-Time Scaling for Vision-Language Models,Fawaz Sammani | Tzoulio Chamiti | Nikos Deligiannis,cs.CV,2026-06-27T11:12:42Z,v1,https://arxiv.org/abs/2606.28864v2
|
||||
2505.15508v1,Multilingual Test-Time Scaling via Initial Thought Transfer,Prasoon Bajpai | Tanmoy Chakraborty,cs.CL,2025-05-21T13:27:38Z,v1,https://arxiv.org/abs/2505.15508v1
|
||||
2501.19393v3,s1: Simple test-time scaling,Niklas Muennighoff | Zitong Yang | Weijia Shi | Xiang Lisa Li | Li Fei-Fei | Hannaneh Hajishirzi | Luke Zettlemoyer | Percy Liang | Emmanuel Candès | Tatsunori Hashimoto,cs.CL; cs.AI; cs.LG,2025-01-31T18:48:08Z,v1,https://arxiv.org/abs/2501.19393v3
|
||||
2508.03333v2,CTTS: Collective Test-Time Scaling,Zhende Song | Shengji Tang | Peng Ye | Jiayuan Fan | Lei Bai | Tao Chen | Wanli Ouyang,cs.CL; cs.AI,2025-08-05T11:19:08Z,v1,https://arxiv.org/abs/2508.03333v2
|
||||
2502.17407v2,Linguistic Generalizability of Test-Time Scaling in Mathematical Reasoning,Guijin Son | Jiwoo Hong | Hyunwoo Ko | James Thorne,cs.CL,2025-02-24T18:36:15Z,v1,https://arxiv.org/abs/2502.17407v2
|
||||
2502.12215v2,Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?,Zhiyuan Zeng | Qinyuan Cheng | Zhangyue Yin | Yunhua Zhou | Xipeng Qiu,cs.LG; cs.AI; cs.CL,2025-02-17T07:21:11Z,v1,https://arxiv.org/abs/2502.12215v2
|
||||
2603.10960v1,Ranking Reasoning LLMs under Test-Time Scaling,Mohsen Hariri | Michael Hinczewski | Jing Ma | Vipin Chaudhary,cs.LG; math.ST,2026-03-11T16:47:41Z,v1,https://arxiv.org/abs/2603.10960v1
|
||||
2605.20473v1,Code Generation by Differential Test Time Scaling,Yifeng He | Ethan Wang | Jicheng Wang | Xuanxin Ouyang | Hao Chen,cs.SE; cs.AI; cs.LG,2026-05-19T20:39:14Z,v1,https://arxiv.org/abs/2605.20473v1
|
||||
2503.24320v2,Can Test-Time Scaling Improve World Foundation Model?,Wenyan Cong | Hanqing Zhu | Peihao Wang | Bangya Liu | Dejia Xu | Kevin Wang | David Z. Pan | Yan Wang | Zhiwen Fan | Zhangyang Wang,cs.CV,2025-03-31T17:07:37Z,v1,https://arxiv.org/abs/2503.24320v2
|
||||
2608.10928v1,ThinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scaling,Vaibhav Singh | Soumya Suvra Ghosal | Sarvesh Gharat | Soumyabrata Pal | Ramasuri Narayanam | Dinesh Manocha,cs.AI,2026-08-11T13:58:07Z,v1,https://arxiv.org/abs/2608.10928v1
|
||||
2509.04474v1,"Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling",Shengyin Sun | Yiming Li | Xing Li | Yingzhao Lian | Weizhe Lin | Hui-Ling Zhen | Zhiyuan Yang | Chen Chen | Xianzhi Yu | Mingxuan Yuan | Chen Ma,cs.CL; cs.AI,2025-08-30T01:54:55Z,v1,https://arxiv.org/abs/2509.04474v1
|
||||
2605.08000v1,Rethinking Dense Optical Flow without Test-Time Scaling,Praroop Chanda | Suryansh Kumar,cs.CV,2026-05-08T16:56:24Z,v1,https://arxiv.org/abs/2605.08000v1
|
||||
2511.15738v2,"Extending Test-Time Scaling: A 3D Perspective with Context, Batch, and Turn",Chao Yu | Qixin Tan | Jiaxuan Gao | Shi Yu | Hong Lu | Xinting Yang | Zelai Xu | Yu Wang | Yi Wu | Eugene Vinitsky,cs.LG; cs.AI,2025-11-18T14:07:57Z,v1,https://arxiv.org/abs/2511.15738v2
|
||||
2509.15148v3,ATTS: Asynchronous Test-Time Scaling via Conformal Prediction,Jing Xiong | Qiujiang Chen | Fanghua Ye | Zhongwei Wan | Chuanyang Zheng | Chenyang Zhao | Hui Shen | Hanbo Li | Chaofan Tao | Haochen Tan | Haoli Bai | Lifeng Shang | Lingpeng Kong | Ngai Wong,cs.CL,2025-09-18T16:55:09Z,v1,https://arxiv.org/abs/2509.15148v3
|
||||
2601.16175v2,Learning to Discover at Test Time,Mert Yuksekgonul | Daniel Koceja | Xinhao Li | Federico Bianchi | Jed McCaleb | Xiaolong Wang | Jan Kautz | Yejin Choi | James Zou | Carlos Guestrin | Yu Sun,cs.LG; cs.AI,2026-01-22T18:24:00Z,v1,https://arxiv.org/abs/2601.16175v2
|
||||
2507.14419v1,It's Not That Simple. An Analysis of Simple Test-Time Scaling,Guojun Wu,cs.LG; cs.AI; cs.CL,2025-07-19T00:28:10Z,v1,https://arxiv.org/abs/2507.14419v1
|
||||
2509.20368v1,LATTS: Locally Adaptive Test-Time Scaling,Theo Uscidda | Matthew Trager | Michael Kleinman | Aditya Chattopadhyay | Wei Xia | Stefano Soatto,cs.AI,2025-09-16T17:51:33Z,v1,https://arxiv.org/abs/2509.20368v1
|
||||
2602.12279v2,UniT: Unified Multimodal Chain-of-Thought Test-time Scaling,Leon Liangyu Chen | Haoyu Ma | Zhipeng Fan | Ziqi Huang | Animesh Sinha | Xiaoliang Dai | Jialiang Wang | Zecheng He | Jianwei Yang | Chunyuan Li | Junzhe Sun | Chu Wang | Serena Yeung-Levy | Felix Juefei-Xu,cs.CV; cs.AI; cs.LG,2026-02-12T18:59:49Z,v1,https://arxiv.org/abs/2602.12279v2
|
||||
|
@@ -0,0 +1,424 @@
|
||||
{
|
||||
"query": "all:\"test-time\" OR all:\"test-time scaling\" OR all:\"test_time compute\"",
|
||||
"count": 20,
|
||||
"fetched": 20,
|
||||
"source": "https://export.arxiv.org/api/query",
|
||||
"papers": [
|
||||
{
|
||||
"arxiv_id": "2608.11829v3",
|
||||
"title": "Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2608.11829v3",
|
||||
"authors": [
|
||||
"Xinmu Ge",
|
||||
"Zizhuo Zhang",
|
||||
"Yu Huang",
|
||||
"Jianing Zhu",
|
||||
"Lin Yuan",
|
||||
"Wanli Gu",
|
||||
"Weichang Wu",
|
||||
"Weiran Huang",
|
||||
"Bo Han",
|
||||
"Xiaolu Zhang",
|
||||
"Jiangchao Yao"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"cs.CL"
|
||||
],
|
||||
"submitted": "2026-08-12T09:13:35Z",
|
||||
"updated": "2026-09-28T12:56:13Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2609.19499v2",
|
||||
"title": "Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2609.19499v2",
|
||||
"authors": [
|
||||
"Mobina Kashaniyan",
|
||||
"Ali Jannesari"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"cs.DC",
|
||||
"cs.PF"
|
||||
],
|
||||
"submitted": "2026-09-16T23:29:57Z",
|
||||
"updated": "2026-09-19T17:04:07Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2606.28864v2",
|
||||
"title": "On Test-Time Scaling for Vision-Language Models",
|
||||
"url": "https://arxiv.org/abs/2606.28864v2",
|
||||
"authors": [
|
||||
"Fawaz Sammani",
|
||||
"Tzoulio Chamiti",
|
||||
"Nikos Deligiannis"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CV"
|
||||
],
|
||||
"submitted": "2026-06-27T11:12:42Z",
|
||||
"updated": "2026-07-01T18:10:39Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2505.15508v1",
|
||||
"title": "Multilingual Test-Time Scaling via Initial Thought Transfer",
|
||||
"url": "https://arxiv.org/abs/2505.15508v1",
|
||||
"authors": [
|
||||
"Prasoon Bajpai",
|
||||
"Tanmoy Chakraborty"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CL"
|
||||
],
|
||||
"submitted": "2025-05-21T13:27:38Z",
|
||||
"updated": "2025-05-21T13:27:38Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2501.19393v3",
|
||||
"title": "s1: Simple test-time scaling",
|
||||
"url": "https://arxiv.org/abs/2501.19393v3",
|
||||
"authors": [
|
||||
"Niklas Muennighoff",
|
||||
"Zitong Yang",
|
||||
"Weijia Shi",
|
||||
"Xiang Lisa Li",
|
||||
"Li Fei-Fei",
|
||||
"Hannaneh Hajishirzi",
|
||||
"Luke Zettlemoyer",
|
||||
"Percy Liang",
|
||||
"Emmanuel Candès",
|
||||
"Tatsunori Hashimoto"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CL",
|
||||
"cs.AI",
|
||||
"cs.LG"
|
||||
],
|
||||
"submitted": "2025-01-31T18:48:08Z",
|
||||
"updated": "2025-03-01T06:07:39Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2508.03333v2",
|
||||
"title": "CTTS: Collective Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2508.03333v2",
|
||||
"authors": [
|
||||
"Zhende Song",
|
||||
"Shengji Tang",
|
||||
"Peng Ye",
|
||||
"Jiayuan Fan",
|
||||
"Lei Bai",
|
||||
"Tao Chen",
|
||||
"Wanli Ouyang"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CL",
|
||||
"cs.AI"
|
||||
],
|
||||
"submitted": "2025-08-05T11:19:08Z",
|
||||
"updated": "2025-09-28T04:43:05Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2502.17407v2",
|
||||
"title": "Linguistic Generalizability of Test-Time Scaling in Mathematical Reasoning",
|
||||
"url": "https://arxiv.org/abs/2502.17407v2",
|
||||
"authors": [
|
||||
"Guijin Son",
|
||||
"Jiwoo Hong",
|
||||
"Hyunwoo Ko",
|
||||
"James Thorne"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CL"
|
||||
],
|
||||
"submitted": "2025-02-24T18:36:15Z",
|
||||
"updated": "2025-08-01T10:09:29Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2502.12215v2",
|
||||
"title": "Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?",
|
||||
"url": "https://arxiv.org/abs/2502.12215v2",
|
||||
"authors": [
|
||||
"Zhiyuan Zeng",
|
||||
"Qinyuan Cheng",
|
||||
"Zhangyue Yin",
|
||||
"Yunhua Zhou",
|
||||
"Xipeng Qiu"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"cs.AI",
|
||||
"cs.CL"
|
||||
],
|
||||
"submitted": "2025-02-17T07:21:11Z",
|
||||
"updated": "2025-03-03T15:29:43Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2603.10960v1",
|
||||
"title": "Ranking Reasoning LLMs under Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2603.10960v1",
|
||||
"authors": [
|
||||
"Mohsen Hariri",
|
||||
"Michael Hinczewski",
|
||||
"Jing Ma",
|
||||
"Vipin Chaudhary"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"math.ST"
|
||||
],
|
||||
"submitted": "2026-03-11T16:47:41Z",
|
||||
"updated": "2026-03-11T16:47:41Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2605.20473v1",
|
||||
"title": "Code Generation by Differential Test Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2605.20473v1",
|
||||
"authors": [
|
||||
"Yifeng He",
|
||||
"Ethan Wang",
|
||||
"Jicheng Wang",
|
||||
"Xuanxin Ouyang",
|
||||
"Hao Chen"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.SE",
|
||||
"cs.AI",
|
||||
"cs.LG"
|
||||
],
|
||||
"submitted": "2026-05-19T20:39:14Z",
|
||||
"updated": "2026-05-19T20:39:14Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2503.24320v2",
|
||||
"title": "Can Test-Time Scaling Improve World Foundation Model?",
|
||||
"url": "https://arxiv.org/abs/2503.24320v2",
|
||||
"authors": [
|
||||
"Wenyan Cong",
|
||||
"Hanqing Zhu",
|
||||
"Peihao Wang",
|
||||
"Bangya Liu",
|
||||
"Dejia Xu",
|
||||
"Kevin Wang",
|
||||
"David Z. Pan",
|
||||
"Yan Wang",
|
||||
"Zhiwen Fan",
|
||||
"Zhangyang Wang"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CV"
|
||||
],
|
||||
"submitted": "2025-03-31T17:07:37Z",
|
||||
"updated": "2025-08-08T17:55:21Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2608.10928v1",
|
||||
"title": "ThinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2608.10928v1",
|
||||
"authors": [
|
||||
"Vaibhav Singh",
|
||||
"Soumya Suvra Ghosal",
|
||||
"Sarvesh Gharat",
|
||||
"Soumyabrata Pal",
|
||||
"Ramasuri Narayanam",
|
||||
"Dinesh Manocha"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.AI"
|
||||
],
|
||||
"submitted": "2026-08-11T13:58:07Z",
|
||||
"updated": "2026-08-11T13:58:07Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2509.04474v1",
|
||||
"title": "Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2509.04474v1",
|
||||
"authors": [
|
||||
"Shengyin Sun",
|
||||
"Yiming Li",
|
||||
"Xing Li",
|
||||
"Yingzhao Lian",
|
||||
"Weizhe Lin",
|
||||
"Hui-Ling Zhen",
|
||||
"Zhiyuan Yang",
|
||||
"Chen Chen",
|
||||
"Xianzhi Yu",
|
||||
"Mingxuan Yuan",
|
||||
"Chen Ma"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CL",
|
||||
"cs.AI"
|
||||
],
|
||||
"submitted": "2025-08-30T01:54:55Z",
|
||||
"updated": "2025-08-30T01:54:55Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2605.08000v1",
|
||||
"title": "Rethinking Dense Optical Flow without Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2605.08000v1",
|
||||
"authors": [
|
||||
"Praroop Chanda",
|
||||
"Suryansh Kumar"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CV"
|
||||
],
|
||||
"submitted": "2026-05-08T16:56:24Z",
|
||||
"updated": "2026-05-08T16:56:24Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2511.15738v2",
|
||||
"title": "Extending Test-Time Scaling: A 3D Perspective with Context, Batch, and Turn",
|
||||
"url": "https://arxiv.org/abs/2511.15738v2",
|
||||
"authors": [
|
||||
"Chao Yu",
|
||||
"Qixin Tan",
|
||||
"Jiaxuan Gao",
|
||||
"Shi Yu",
|
||||
"Hong Lu",
|
||||
"Xinting Yang",
|
||||
"Zelai Xu",
|
||||
"Yu Wang",
|
||||
"Yi Wu",
|
||||
"Eugene Vinitsky"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"cs.AI"
|
||||
],
|
||||
"submitted": "2025-11-18T14:07:57Z",
|
||||
"updated": "2025-11-21T15:10:10Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2509.15148v3",
|
||||
"title": "ATTS: Asynchronous Test-Time Scaling via Conformal Prediction",
|
||||
"url": "https://arxiv.org/abs/2509.15148v3",
|
||||
"authors": [
|
||||
"Jing Xiong",
|
||||
"Qiujiang Chen",
|
||||
"Fanghua Ye",
|
||||
"Zhongwei Wan",
|
||||
"Chuanyang Zheng",
|
||||
"Chenyang Zhao",
|
||||
"Hui Shen",
|
||||
"Hanbo Li",
|
||||
"Chaofan Tao",
|
||||
"Haochen Tan",
|
||||
"Haoli Bai",
|
||||
"Lifeng Shang",
|
||||
"Lingpeng Kong",
|
||||
"Ngai Wong"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CL"
|
||||
],
|
||||
"submitted": "2025-09-18T16:55:09Z",
|
||||
"updated": "2026-02-24T08:46:07Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2601.16175v2",
|
||||
"title": "Learning to Discover at Test Time",
|
||||
"url": "https://arxiv.org/abs/2601.16175v2",
|
||||
"authors": [
|
||||
"Mert Yuksekgonul",
|
||||
"Daniel Koceja",
|
||||
"Xinhao Li",
|
||||
"Federico Bianchi",
|
||||
"Jed McCaleb",
|
||||
"Xiaolong Wang",
|
||||
"Jan Kautz",
|
||||
"Yejin Choi",
|
||||
"James Zou",
|
||||
"Carlos Guestrin",
|
||||
"Yu Sun"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"cs.AI"
|
||||
],
|
||||
"submitted": "2026-01-22T18:24:00Z",
|
||||
"updated": "2026-02-05T18:03:03Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2507.14419v1",
|
||||
"title": "It's Not That Simple. An Analysis of Simple Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2507.14419v1",
|
||||
"authors": [
|
||||
"Guojun Wu"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.LG",
|
||||
"cs.AI",
|
||||
"cs.CL"
|
||||
],
|
||||
"submitted": "2025-07-19T00:28:10Z",
|
||||
"updated": "2025-07-19T00:28:10Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2509.20368v1",
|
||||
"title": "LATTS: Locally Adaptive Test-Time Scaling",
|
||||
"url": "https://arxiv.org/abs/2509.20368v1",
|
||||
"authors": [
|
||||
"Theo Uscidda",
|
||||
"Matthew Trager",
|
||||
"Michael Kleinman",
|
||||
"Aditya Chattopadhyay",
|
||||
"Wei Xia",
|
||||
"Stefano Soatto"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.AI"
|
||||
],
|
||||
"submitted": "2025-09-16T17:51:33Z",
|
||||
"updated": "2025-09-16T17:51:33Z",
|
||||
"source": "v1"
|
||||
},
|
||||
{
|
||||
"arxiv_id": "2602.12279v2",
|
||||
"title": "UniT: Unified Multimodal Chain-of-Thought Test-time Scaling",
|
||||
"url": "https://arxiv.org/abs/2602.12279v2",
|
||||
"authors": [
|
||||
"Leon Liangyu Chen",
|
||||
"Haoyu Ma",
|
||||
"Zhipeng Fan",
|
||||
"Ziqi Huang",
|
||||
"Animesh Sinha",
|
||||
"Xiaoliang Dai",
|
||||
"Jialiang Wang",
|
||||
"Zecheng He",
|
||||
"Jianwei Yang",
|
||||
"Chunyuan Li",
|
||||
"Junzhe Sun",
|
||||
"Chu Wang",
|
||||
"Serena Yeung-Levy",
|
||||
"Felix Juefei-Xu"
|
||||
],
|
||||
"subjects": [
|
||||
"cs.CV",
|
||||
"cs.AI",
|
||||
"cs.LG"
|
||||
],
|
||||
"submitted": "2026-02-12T18:59:49Z",
|
||||
"updated": "2026-06-12T18:20:54Z",
|
||||
"source": "v1"
|
||||
}
|
||||
]
|
||||
}
|
||||
@@ -0,0 +1,120 @@
|
||||
1. id=2608.11829v3 title=Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling
|
||||
authors: Xinmu Ge, Zizhuo Zhang, Yu Huang, Jianing Zhu, Lin Yuan, Wanli Gu, Weichang Wu, Weiran Huang, Bo Han, Xiaolu Zhang, Jiangchao Yao
|
||||
subjects: cs.LG, cs.CL submitted: 2026-08-12T09:13:35Z source: v1
|
||||
url: https://arxiv.org/abs/2608.11829v3
|
||||
source: v1
|
||||
|
||||
2. id=2609.19499v2 title=Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling
|
||||
authors: Mobina Kashaniyan, Ali Jannesari
|
||||
subjects: cs.LG, cs.DC, cs.PF submitted: 2026-09-16T23:29:57Z source: v1
|
||||
url: https://arxiv.org/abs/2609.19499v2
|
||||
source: v1
|
||||
|
||||
3. id=2606.28864v2 title=On Test-Time Scaling for Vision-Language Models
|
||||
authors: Fawaz Sammani, Tzoulio Chamiti, Nikos Deligiannis
|
||||
subjects: cs.CV submitted: 2026-06-27T11:12:42Z source: v1
|
||||
url: https://arxiv.org/abs/2606.28864v2
|
||||
source: v1
|
||||
|
||||
4. id=2505.15508v1 title=Multilingual Test-Time Scaling via Initial Thought Transfer
|
||||
authors: Prasoon Bajpai, Tanmoy Chakraborty
|
||||
subjects: cs.CL submitted: 2025-05-21T13:27:38Z source: v1
|
||||
url: https://arxiv.org/abs/2505.15508v1
|
||||
source: v1
|
||||
|
||||
5. id=2501.19393v3 title=s1: Simple test-time scaling
|
||||
authors: Niklas Muennighoff, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel Candès, Tatsunori Hashimoto
|
||||
subjects: cs.CL, cs.AI, cs.LG submitted: 2025-01-31T18:48:08Z source: v1
|
||||
url: https://arxiv.org/abs/2501.19393v3
|
||||
source: v1
|
||||
|
||||
6. id=2508.03333v2 title=CTTS: Collective Test-Time Scaling
|
||||
authors: Zhende Song, Shengji Tang, Peng Ye, Jiayuan Fan, Lei Bai, Tao Chen, Wanli Ouyang
|
||||
subjects: cs.CL, cs.AI submitted: 2025-08-05T11:19:08Z source: v1
|
||||
url: https://arxiv.org/abs/2508.03333v2
|
||||
source: v1
|
||||
|
||||
7. id=2502.17407v2 title=Linguistic Generalizability of Test-Time Scaling in Mathematical Reasoning
|
||||
authors: Guijin Son, Jiwoo Hong, Hyunwoo Ko, James Thorne
|
||||
subjects: cs.CL submitted: 2025-02-24T18:36:15Z source: v1
|
||||
url: https://arxiv.org/abs/2502.17407v2
|
||||
source: v1
|
||||
|
||||
8. id=2502.12215v2 title=Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?
|
||||
authors: Zhiyuan Zeng, Qinyuan Cheng, Zhangyue Yin, Yunhua Zhou, Xipeng Qiu
|
||||
subjects: cs.LG, cs.AI, cs.CL submitted: 2025-02-17T07:21:11Z source: v1
|
||||
url: https://arxiv.org/abs/2502.12215v2
|
||||
source: v1
|
||||
|
||||
9. id=2603.10960v1 title=Ranking Reasoning LLMs under Test-Time Scaling
|
||||
authors: Mohsen Hariri, Michael Hinczewski, Jing Ma, Vipin Chaudhary
|
||||
subjects: cs.LG, math.ST submitted: 2026-03-11T16:47:41Z source: v1
|
||||
url: https://arxiv.org/abs/2603.10960v1
|
||||
source: v1
|
||||
|
||||
10. id=2605.20473v1 title=Code Generation by Differential Test Time Scaling
|
||||
authors: Yifeng He, Ethan Wang, Jicheng Wang, Xuanxin Ouyang, Hao Chen
|
||||
subjects: cs.SE, cs.AI, cs.LG submitted: 2026-05-19T20:39:14Z source: v1
|
||||
url: https://arxiv.org/abs/2605.20473v1
|
||||
source: v1
|
||||
|
||||
11. id=2503.24320v2 title=Can Test-Time Scaling Improve World Foundation Model?
|
||||
authors: Wenyan Cong, Hanqing Zhu, Peihao Wang, Bangya Liu, Dejia Xu, Kevin Wang, David Z. Pan, Yan Wang, Zhiwen Fan, Zhangyang Wang
|
||||
subjects: cs.CV submitted: 2025-03-31T17:07:37Z source: v1
|
||||
url: https://arxiv.org/abs/2503.24320v2
|
||||
source: v1
|
||||
|
||||
12. id=2608.10928v1 title=ThinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scaling
|
||||
authors: Vaibhav Singh, Soumya Suvra Ghosal, Sarvesh Gharat, Soumyabrata Pal, Ramasuri Narayanam, Dinesh Manocha
|
||||
subjects: cs.AI submitted: 2026-08-11T13:58:07Z source: v1
|
||||
url: https://arxiv.org/abs/2608.10928v1
|
||||
source: v1
|
||||
|
||||
13. id=2509.04474v1 title=Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling
|
||||
authors: Shengyin Sun, Yiming Li, Xing Li, Yingzhao Lian, Weizhe Lin, Hui-Ling Zhen, Zhiyuan Yang, Chen Chen, Xianzhi Yu, Mingxuan Yuan, Chen Ma
|
||||
subjects: cs.CL, cs.AI submitted: 2025-08-30T01:54:55Z source: v1
|
||||
url: https://arxiv.org/abs/2509.04474v1
|
||||
source: v1
|
||||
|
||||
14. id=2605.08000v1 title=Rethinking Dense Optical Flow without Test-Time Scaling
|
||||
authors: Praroop Chanda, Suryansh Kumar
|
||||
subjects: cs.CV submitted: 2026-05-08T16:56:24Z source: v1
|
||||
url: https://arxiv.org/abs/2605.08000v1
|
||||
source: v1
|
||||
|
||||
15. id=2511.15738v2 title=Extending Test-Time Scaling: A 3D Perspective with Context, Batch, and Turn
|
||||
authors: Chao Yu, Qixin Tan, Jiaxuan Gao, Shi Yu, Hong Lu, Xinting Yang, Zelai Xu, Yu Wang, Yi Wu, Eugene Vinitsky
|
||||
subjects: cs.LG, cs.AI submitted: 2025-11-18T14:07:57Z source: v1
|
||||
url: https://arxiv.org/abs/2511.15738v2
|
||||
source: v1
|
||||
|
||||
16. id=2509.15148v3 title=ATTS: Asynchronous Test-Time Scaling via Conformal Prediction
|
||||
authors: Jing Xiong, Qiujiang Chen, Fanghua Ye, Zhongwei Wan, Chuanyang Zheng, Chenyang Zhao, Hui Shen, Hanbo Li, Chaofan Tao, Haochen Tan, Haoli Bai, Lifeng Shang, Lingpeng Kong, Ngai Wong
|
||||
subjects: cs.CL submitted: 2025-09-18T16:55:09Z source: v1
|
||||
url: https://arxiv.org/abs/2509.15148v3
|
||||
source: v1
|
||||
|
||||
17. id=2601.16175v2 title=Learning to Discover at Test Time
|
||||
authors: Mert Yuksekgonul, Daniel Koceja, Xinhao Li, Federico Bianchi, Jed McCaleb, Xiaolong Wang, Jan Kautz, Yejin Choi, James Zou, Carlos Guestrin, Yu Sun
|
||||
subjects: cs.LG, cs.AI submitted: 2026-01-22T18:24:00Z source: v1
|
||||
url: https://arxiv.org/abs/2601.16175v2
|
||||
source: v1
|
||||
|
||||
18. id=2507.14419v1 title=It's Not That Simple. An Analysis of Simple Test-Time Scaling
|
||||
authors: Guojun Wu
|
||||
subjects: cs.LG, cs.AI, cs.CL submitted: 2025-07-19T00:28:10Z source: v1
|
||||
url: https://arxiv.org/abs/2507.14419v1
|
||||
source: v1
|
||||
|
||||
19. id=2509.20368v1 title=LATTS: Locally Adaptive Test-Time Scaling
|
||||
authors: Theo Uscidda, Matthew Trager, Michael Kleinman, Aditya Chattopadhyay, Wei Xia, Stefano Soatto
|
||||
subjects: cs.AI submitted: 2025-09-16T17:51:33Z source: v1
|
||||
url: https://arxiv.org/abs/2509.20368v1
|
||||
source: v1
|
||||
|
||||
20. id=2602.12279v2 title=UniT: Unified Multimodal Chain-of-Thought Test-time Scaling
|
||||
authors: Leon Liangyu Chen, Haoyu Ma, Zhipeng Fan, Ziqi Huang, Animesh Sinha, Xiaoliang Dai, Jialiang Wang, Zecheng He, Jianwei Yang, Chunyuan Li, Junzhe Sun, Chu Wang, Serena Yeung-Levy, Felix Juefei-Xu
|
||||
subjects: cs.CV, cs.AI, cs.LG submitted: 2026-02-12T18:59:49Z source: v1
|
||||
url: https://arxiv.org/abs/2602.12279v2
|
||||
source: v1
|
||||
|
||||
Reference in new issue
Block a user