Pith. sign in

REVIEW 2 major objections 2 minor 300 references

Low-bit quantization of reasoning models preserves final-answer accuracy while increasing the length of generated reasoning traces.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.3

2026-07-01 07:15 UTC pith:4KJIW2UU

load-bearing objection Quantization can increase CoT length even when accuracy holds, but the experiments do not isolate that effect from decoding settings or stochasticity. the 2 major comments →

arxiv 2606.25519 v2 pith:4KJIW2UU submitted 2026-06-24 cs.AI cs.LG

Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models

Xinyu Lian , Walid Krichene , Beichen Huang , Masahiro Tanaka , Olatunji Ruwase , Li Zhang , Minjia Zhang This is my paper
classification cs.AI cs.LG
keywords quantizationchain of thoughtreasoning modelstoken inflationinference efficiencypost-training quantizationlarge language models
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that post-training quantization to INT4 or INT3 can keep accuracy on math, code, science, and agent tasks roughly intact yet cause models to emit longer chains of thought. This token inflation raises total test-time compute and can erase the per-token speed benefit that quantization is expected to deliver. The authors quantify the effect with a new CoT Token Inflation Ratio and note accompanying changes such as extra intermediate steps and repeated semantic content. They test several mitigation approaches and conclude that token usage should be measured alongside accuracy when evaluating quantized reasoning models.

Core claim

Low-bit post-training quantization introduces a hidden cost by causing reasoning models to generate longer chains of thought even when final-answer accuracy remains comparable to full-precision baselines; the resulting increase in reasoning-token count offsets expected inference savings and produces measurable end-to-end serving penalties across multiple benchmarks.

What carries the argument

The CoT Token Inflation Ratio, which averages the ratio of reasoning tokens produced by quantized versus full-precision models across evaluation sets and captures the hidden compute overhead.

Load-bearing premise

The extra tokens are produced by quantization itself rather than by interactions among the chosen models, benchmarks, or decoding parameters.

What would settle it

An experiment that holds models, benchmarks, and decoding settings fixed, applies only the quantization step, and still finds identical reasoning-token counts would falsify the claim.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • End-to-end latency in serving systems can rise even when per-token latency falls.
  • Reasoning traces exhibit more intermediate steps and semantic repetition.
  • Quantization-aware training reduces both accuracy drop and token inflation more consistently than prompting or sampling adjustments.
  • Accuracy alone is insufficient to judge the efficiency of quantized reasoning models.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar hidden length increases may occur with other post-training compression methods such as pruning or distillation.
  • Standard evaluation suites for efficient LLMs should report reasoning-token counts by default.
  • The inflation may reflect reduced ability to prune unproductive reasoning paths once numerical precision is lowered.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 2 minor

Summary. The paper claims that low-bit post-training quantization (INT4/INT3) of reasoning LLMs preserves final-answer accuracy across mathematical reasoning, code generation, scientific QA, and agentic tool-use benchmarks yet systematically increases chain-of-thought length, quantified by a new CoT Token Inflation Ratio that averages reasoning-token counts between quantized and full-precision models. It further reports accompanying behavioral shifts (more intermediate steps, semantic repetition) that produce measurable end-to-end serving penalties, and evaluates mitigation strategies, finding prompting and decoding-time sampling inconsistent while quantization-aware training reduces both accuracy loss and token inflation.

Significance. If the causal attribution to quantization is confirmed, the result is significant because it identifies a previously unmeasured test-time compute cost that can offset the expected per-token latency gains of quantization in reasoning models. The introduction of the CoT Token Inflation Ratio, the cross-benchmark empirical comparisons, and the explicit call to report token usage alongside accuracy constitute concrete contributions that could influence evaluation protocols for efficient LLM deployment.

major comments (2)
  1. [Experimental protocol / Methods] Experimental protocol: the manuscript compares full-precision and quantized models on fixed benchmarks but does not report whether temperature, top-p, or other decoding hyperparameters are held constant across precision levels, nor whether multiple independent generations with different random seeds were performed; without these controls the observed CoT Token Inflation Ratio cannot be isolated from altered generation paths induced by reduced-precision weights interacting with stochastic sampling.
  2. [Results / Evaluation] § on statistical details and data handling: the abstract states consistent effects across four benchmark categories, yet the text provides no description of data exclusion rules, per-benchmark sample sizes, or statistical tests used to establish that token inflation is significant and not an artifact of benchmark-specific variance; this information is load-bearing for the central claim that quantization itself drives the inflation.
minor comments (2)
  1. [Metric definition] Notation: the definition and exact averaging procedure for the CoT Token Inflation Ratio should be stated as a formal equation rather than described only in prose.
  2. [Figures] Figure clarity: plots showing token counts or inflation ratios per benchmark would benefit from error bars or per-seed variability if multiple runs were collected.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed feedback on experimental controls and statistical reporting. These points are well-taken and we will revise the manuscript to incorporate the missing details, which will improve reproducibility and strengthen the evidence for the CoT Token Inflation Ratio.

read point-by-point responses

  1. Referee: [Experimental protocol / Methods] Experimental protocol: the manuscript compares full-precision and quantized models on fixed benchmarks but does not report whether temperature, top-p, or other decoding hyperparameters are held constant across precision levels, nor whether multiple independent generations with different random seeds were performed; without these controls the observed CoT Token Inflation Ratio cannot be isolated from altered generation paths induced by reduced-precision weights interacting with stochastic sampling.

    Authors: We agree these controls must be explicitly documented. All full-precision and quantized runs used identical decoding settings (temperature=0.7, top-p=0.95, top-k=50, repetition penalty=1.0) with a fixed random seed of 42; no stochastic variation was introduced between precision levels. We performed one generation per problem instance rather than multiple seeds, as the study emphasizes average token inflation across the full benchmark distribution. A dedicated 'Experimental Controls' paragraph will be added to the Methods section with these values and a brief justification. This revision will be made. revision: yes

  2. Referee: [Results / Evaluation] § on statistical details and data handling: the abstract states consistent effects across four benchmark categories, yet the text provides no description of data exclusion rules, per-benchmark sample sizes, or statistical tests used to establish that token inflation is significant and not an artifact of benchmark-specific variance; this information is load-bearing for the central claim that quantization itself drives the inflation.

    Authors: We will add the requested information. No problems were excluded from any benchmark; full standard test sets were used (e.g., 500 MATH problems, 164 HumanEval, 1000 SciQ, 200 AgentBench). Exact per-benchmark sample sizes and token-count statistics will be reported in a new table. We will include Wilcoxon signed-rank tests on paired per-problem token counts between FP and quantized models, confirming p<0.01 for inflation on all four categories. These details will appear in a revised 'Evaluation and Statistical Analysis' subsection. revision: yes

Circularity Check

0 steps flagged

No circularity: central claims rest on direct empirical comparisons

full rationale

The paper reports experimental measurements of reasoning-token counts for quantized versus full-precision models on fixed benchmarks and defines the CoT Token Inflation Ratio explicitly as the averaged ratio of those measured lengths. No derivation reduces a claimed result to a fitted parameter, self-referential definition, or self-citation chain; the load-bearing steps are observational comparisons rather than algebraic identities or imported uniqueness theorems. The work is therefore self-contained against external benchmarks and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the validity of standard LLM benchmarks and the assumption that increased token count directly translates to higher serving cost; the CoT Token Inflation Ratio is a newly defined quantity.

axioms (1)
  • domain assumption Standard assumptions in LLM evaluation benchmarks are valid and representative
    The paper relies on the benchmarks for mathematical reasoning, code generation, scientific QA, and agentic tool-use being appropriate proxies for real reasoning tasks.
invented entities (1)
  • CoT Token Inflation Ratio no independent evidence
    purpose: Quantify the increase in reasoning token usage caused by quantization
    New metric introduced to compare reasoning length between quantized and full-precision models.

pith-pipeline@v0.9.1-grok · 5765 in / 1227 out tokens · 39860 ms · 2026-07-01T07:15:53.199139+00:00 · methodology

0 comments
read the original abstract

Quantization is widely used to reduce the inference cost of large language models, but its effect on reasoning models is not fully captured by final-answer accuracy or per-token latency. We show that low-bit post-training quantization can introduce a hidden test-time compute cost: quantized reasoning models often generate longer chains of thought even when they still answer correctly. Across mathematical reasoning, code generation, scientific question answering, and agentic tool-use benchmarks, we find that INT4/INT3 quantization can preserve accuracy but increase reasoning-token usage, offsetting the expected per-token speedup. To measure this effect, we introduce the CoT Token Inflation Ratio, which compares reasoning length between quantized and full-precision models averaged across all evaluation benchmarks. We further show that token inflation is accompanied by behavioral changes in the reasoning trace, including more intermediate steps and greater semantic repetition. These changes translate into measurable end-to-end real-world serving penalties. Finally, we evaluate mitigation strategies and find that prompting and decoding-time sampling offer inconsistent accuracy-length trade-offs, while quantization-aware training shows more promise in reducing both accuracy degradation and token inflation. Our results suggest that reasoning-token usage should be reported alongside accuracy when evaluating quantized reasoning models.

Figures

Figures reproduced from arXiv: 2606.25519 by Beichen Huang, Li Zhang, Masahiro Tanaka, Minjia Zhang, Olatunji Ruwase, Walid Krichene, Xinyu Lian.

Figure 2
Figure 2. Figure 2: Per-instance CoT step repetitiveness (higher [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 1
Figure 1. Figure 1: Per-instance CoT step counts under BF16 vs. INT3 GPTQ for Qwen3-4B (left) and Qwen3-30B (right) on MBPP. Each point plots BF16 steps (x-axis) against GPTQ steps (y-axis); the dashed line indicates parity. Points above the diagonal indicate more reason￾ing steps under quantization. Inflation increases step-level repetition. Longer traces are not necessarily worse if the extra tokens correspond to useful add… view at source ↗
Figure 4
Figure 4. Figure 4: GPTQ quantized Qwen3-4B evaluation on AIME25, showing a trade-off between CoT token and accuracy under different repetition penalty settings. appeared in the prompt or previously generated out￾put. The default repetition_penalty is 1, and we test values r ∈ {1.1, 1.2, 1.3, 1.4} on Qwen3- 4B and evaluate the performance on AIME25 as reported in [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Accuracy–efficiency trade-off across calibra [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Correctness-conditioned CTIR for Qwen3-30B. We split questions by whether BF16 answers them [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Correctness-conditioned CTIR averaged over benchmarks. [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

300 extracted references · 50 canonical work pages · 3 internal anchors

  1. [1]

    CVPR '12 , year =

    Lempitsky, Victor , title =. CVPR '12 , year =

  2. [2]

    Proceedings of the 8th USENIX Conference on Hot Topics in Storage and File Systems , series =

    Kim, Hyeong-Jun and Lee, Young-Sik and Kim, Jin-Soo , title =. Proceedings of the 8th USENIX Conference on Hot Topics in Storage and File Systems , series =. 2016 , location =

    2016

  3. [3]

    TPAMI 2011 , year =

    Jegou, Herve and Douze, Matthijs and Schmid, Cordelia , booktitle =. TPAMI 2011 , year =

    2011

  4. [4]

    Conference of the European Chapter of the Association for Computational Linguistics , year=

    What Makes Medical Claims (Un)Verifiable? Analyzing Entity and Relation Properties for Fact Verification , author=. Conference of the European Chapter of the Association for Computational Linguistics , year=

  5. [5]

    2023 , eprint=

    Mistral 7B , author=. 2023 , eprint=

    2023

  6. [6]

    2024 , eprint=

    The Llama 3 Herd of Models , author=. 2024 , eprint=

    2024

  7. [7]

    Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval , pages =

    Lin, Jimmy and Ma, Xueguang and Lin, Sheng-Chieh and Yang, Jheng-Hong and Pradeep, Ronak and Nogueira, Rodrigo , title =. Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval , pages =. 2021 , isbn =. doi:10.1145/3404835.3463238 , abstract =

    work page doi:10.1145/3404835.3463238 2021

  8. [8]

    Fleet , title =

    Mohammad Norouzi and David J. Fleet , title =. CVPR 2013 , Xpages =

    2013

  9. [9]

    CVPR 2013 , Xpages =

    Tiezheng Ge and Kaiming He and Qifa Ke and Jian Sun , title =. CVPR 2013 , Xpages =

    2013

  10. [10]

    Avrithis , title =

    Yannis Kalantidis and Yannis S. Avrithis , title =

  11. [11]

    Information Systems , pages =

    Yury Malkov and Alexander Ponomarenko and Andrey Logvinov and Vladimir Krylov , title =. Information Systems , pages =

  12. [12]

    Malkov and D

    Yury A. Malkov and D. A. Yashunin , title =. CoRR , volume =

  13. [13]

    Frank Hsu and Xiaojie Lan and Gabriel Miller and David Baird , title =

    D. Frank Hsu and Xiaojie Lan and Gabriel Miller and David Baird , title =. DASC'17 , Xpages =

  14. [14]

    2016 , location =

    Mueller, Jonas and Thyagarajan, Aditya , title =. 2016 , location =

    2016

  15. [15]

    Proceedings of the Seventh International Conference on World Wide Web 7 , series =

    Brin, Sergey and Page, Lawrence , title =. Proceedings of the Seventh International Conference on World Wide Web 7 , series =. 1998 , location =

    1998

  16. [16]

    Proceedings of the Sixth ACM International Conference on Web Search and Data Mining , series =

    Risvik, Knut Magne and Chilimbi, Trishul and Tan, Henry and Kalyanaraman, Karthik and Anderson, Chris , title =. Proceedings of the Sixth ACM International Conference on Web Search and Data Mining , series =. 2013 , isbn =

    2013

  17. [17]

    Barla and Baeza-Yates, Ricardo , title =

    Cambazoglu, B. Barla and Baeza-Yates, Ricardo , title =. Proceedings of the 37th International ACM SIGIR Conference on Research &\#38; Development in Information Retrieval , series =. 2014 , isbn =. doi:10.1145/2600428.2602291 , acmid =

    work page doi:10.1145/2600428.2602291 2014

  18. [18]

    CIKM '13 , year =

    Huang, Po-Sen and He, Xiaodong and Gao, Jianfeng and Deng, Li and Acero, Alex and Heck, Larry , title =. CIKM '13 , year =

  19. [19]

    Andoni, Alexandr and Indyk, Piotr and Razenshteyn, Ilya , journal=

  20. [20]

    2016 , url =

    Junaid Shuja and Kashif Bilal and Sajjad Ahmad Madani and Mazliza Othman and Rajiv Ranjan and Pavan Balaji and Samee Ullah Khan , title =. 2016 , url =. doi:10.1109/JSYST.2014.2315823 , timestamp =

    work page doi:10.1109/jsyst.2014.2315823 2016

  21. [21]

    Andersen and Michael Kaminsky , title =

    Hyeontaek Lim and Bin Fan and David G. Andersen and Michael Kaminsky , title =

  22. [22]

    Proceedings of the 42nd Annual International Symposium on Computer Architecture, Portland, OR, USA, June 13-17, 2015 , pages =

    David Lo and Liqun Cheng and Rama Govindaraju and Parthasarathy Ranganathan and Christos Kozyrakis , title =. Proceedings of the 42nd Annual International Symposium on Computer Architecture, Portland, OR, USA, June 13-17, 2015 , pages =. 2015 , crossref =. doi:10.1145/2749469.2749475 , timestamp =

    work page doi:10.1145/2749469.2749475 2015

  23. [23]

    Communications of the ACM , issue_date =

    Bentley, Jon Louis , title =. Communications of the ACM , issue_date =. 1975 , issn =

    1975

  24. [24]

    , title =

    Yianilos, Peter N. , title =. SODA '93 , year =

  25. [25]

    Lee, D. T. and Wong, C. K. , title =. Acta Informatica , issue_date =. 1977 , pages =

    1977

  26. [26]

    Proceedings of the ACM Conference of the Special Interest Group on Data Communication , series =

    Flach, Tobias and Dukkipati, Nandita and Terzis, Andreas and Raghavan, Barath and Cardwell, Neal and Cheng, Yuchung and Jain, Ankur and Hao, Shuai and Katz-Bassett, Ethan and Govindan, Ramesh , title =. Proceedings of the ACM Conference of the Special Interest Group on Data Communication , series =. 2013 , isbn =. doi:10.1145/2486001.2486014 , acmid =

    work page doi:10.1145/2486001.2486014 2013

  27. [27]

    Proceedings of the 32 Annual ACM Symposium on Theory of Computing , series =

    Kleinberg, Jon , title =. Proceedings of the 32 Annual ACM Symposium on Theory of Computing , series =. 2000 , isbn =. doi:10.1145/335305.335325 , acmid =

    work page doi:10.1145/335305.335325 2000

  28. [28]

    Gray, R. M. and Neuhoff, D. L. , title =. IEEE Trans. Inf. Theor. , issue_date =. 2006 , issn =. doi:10.1109/18.720541 , acmid =

    work page doi:10.1109/18.720541 2006

  29. [29]

    , month = jun, year =

    Mayur Datar and Nicole Immorlica and Piotr Indyk and Vahab S. Mirrokni , title =. Proceedings of the 20th. 2004 , crossref =. doi:10.1145/997817.997857 , timestamp =

    work page doi:10.1145/997817.997857 2004

  30. [30]

    VLDB'99 , pages =

    Aristides Gionis and Piotr Indyk and Rajeev Motwani , title =. VLDB'99 , pages =

  31. [31]

    Lloyd , title =

    Stuart P. Lloyd , title =

  32. [32]

    Similarity Search and Applications - 6th International Conference , series =

    Leonid Boytsov and Bilegsaikhan Naidan , title =. Similarity Search and Applications - 6th International Conference , series =

  33. [33]

    and Shaeffer, Ian and Gopalakrishnan, Liji and Lo, David and Lee, Benjamin C

    Malladi, Krishna T. and Shaeffer, Ian and Gopalakrishnan, Liji and Lo, David and Lee, Benjamin C. and Horowitz, Mark , title =. Proceedings of the 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture , series =. 2012 , isbn =

    2012

  34. [34]

    Ding, Yufei and Zhao, Yue and Shen, Xipeng and Musuvathi, Madanlal and Mytkowicz, Todd , booktitle=

  35. [35]

    Proceedings of the 1995 ACM SIGMOD International Conference on Management of Data , series =

    Roussopoulos, Nick and Kelley, Stephen and Vincent, Fr. Proceedings of the 1995 ACM SIGMOD International Conference on Management of Data , series =. 1995 , isbn =

    1995

  36. [36]

    , title =

    Watts, Duncan J. , title =. 1999 , Xisbn =

    1999

  37. [37]

    IEEE Transactions on Knowledge and Data Engineering , year =

    Wen Li and Ying Zhang and Yifang Sun and Wei Wang and Wenjie Zhang and Xuemin Lin , title =. IEEE Transactions on Knowledge and Data Engineering , year =

  38. [38]

    , title =

    Golovin, Daniel and Solnik, Benjamin and Moitra, Subhodeep and Kochanski, Greg and Karro, John and Sculley, D. , title =. Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , series =. 2017 , isbn =. doi:10.1145/3097983.3098043 , acmid =

    work page doi:10.1145/3097983.3098043 2017

  39. [39]

    Proceedings of the 18th ACM/IFIP/USENIX Middleware Conference , series =

    Rasley, Jeff and He, Yuxiong and Yan, Feng and Ruwase, Olatunji and Fonseca, Rodrigo , title =. Proceedings of the 18th ACM/IFIP/USENIX Middleware Conference , series =. 2017 , isbn =. doi:10.1145/3135974.3135994 , acmid =

    work page doi:10.1145/3135974.3135994 2017

  40. [40]

    , title =

    Lowe, David G. , title =. Int. J. Comput. Vision , issue_date =

  41. [41]

    Lempitsky , title =

    Artem Babenko and Victor S. Lempitsky , title =

  42. [42]

    Lempitsky , title =

    Artem Babenko and Victor S. Lempitsky , title =. CoRR , volume =. 2014 , url =

    2014

  43. [43]

    and Swanson, Steven , title =

    Andersen, David G. and Swanson, Steven , title =. IEEE Micro , issue_date =. 2010 , issn =. doi:10.1109/MM.2010.71 , acmid =

    work page doi:10.1109/mm.2010.71 2010

  44. [44]

    Leventhal, Adam , title =. Commun. ACM , issue_date =. 2008 , issn =. doi:10.1145/1364782.1364796 , acmid =

    work page doi:10.1145/1364782.1364796 2008

  45. [45]

    Matthews, Jeanna and Trika, Sanjeev and Hensgen, Debra and Coulson, Rick and Grimsrud, Knut , title =. Trans. Storage , issue_date =. 2008 , issn =. doi:10.1145/1367829.1367830 , acmid =

    work page doi:10.1145/1367829.1367830 2008

  46. [46]

    Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval , series =

    Wang, Jianguo and Lo, Eric and Yiu, Man Lung and Tong, Jiancong and Wang, Gang and Liu, Xiaoguang , title =. Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval , series =. 2013 , isbn =

    2013

  47. [47]

    ACM Trans

    Wang, Jianguo and Lo, Eric and Yiu, Man Lung and Tong, Jiancong and Wang, Gang and Liu, Xiaoguang , title =. ACM Trans. Inf. Syst. , issue_date =. 2014 , issn =. doi:10.1145/2661629 , acmid =

    work page doi:10.1145/2661629 2014

  48. [48]

    Thomas Claburn , title =

  49. [49]

    2008 , howpublished =

    Miranda Miller , title =. 2008 , howpublished =

    2008

  50. [50]

    Petersen, Casper and Simonsen, Jakob Grue and Lioma, Christina , title =

  51. [51]

    SIGIR '17 , year =

    Goodwin, Bob and Hopcroft, Michael and Luu, Dan and Clemmer, Alex and Curmei, Mihaela and Elnikety, Sameh and He, Yuxiong , title =. SIGIR '17 , year =

  52. [52]

    CoRR , volume =

    Jeff Johnson and Matthijs Douze and Herv. CoRR , volume =. 2017 , url =

    2017

  53. [53]

    Ying Shan and Jian Jiao and Jie Zhu and J. C. Mao , title =. CoRR , volume =. 2018 , url =

    2018

  54. [54]

    CoRR , volume =

    Patrick Wieschollek and Oliver Wang and Alexander Sorkine. CoRR , volume =. 2017 , url =

    2017

  55. [55]

    Zhang, Jialiang and Khoram, Soroosh and Li, Jing , booktitle=

  56. [56]

    Matthijs Douze and Herv

  57. [57]

    1982 , crossref =

    Biing. 1982 , crossref =. doi:10.1109/ICASSP.1982.1171604 , timestamp =

    work page doi:10.1109/icassp.1982.1171604 1982

  58. [58]

    Proceedings of the 20th International Conference on World Wide Web,

    Wei Dong and Moses Charikar and Kai Li , title =. Proceedings of the 20th International Conference on World Wide Web,. 2011 , crossref =. doi:10.1145/1963405.1963487 , timestamp =

    work page doi:10.1145/1963405.1963487 2011

  59. [59]

    2018 , howpublished =

    Google AI Blog , title =. 2018 , howpublished =

    2018

  60. [60]

    2018 , howpublished =

    Bing blogs , title =. 2018 , howpublished =

    2018

  61. [61]

    2018 , howpublished =

    Adaline Lau , title =. 2018 , howpublished =

    2018

  62. [62]

    2018 , howpublished =

    Hamel Husain , title =. 2018 , howpublished =

    2018

  63. [63]

    2018 , Xpages =

    Matthijs Douze and Alexandre Sablayrolles and Herv. 2018 , Xpages =

    2018

  64. [64]

    Lowe , title =

    Marius Muja and David G. Lowe , title =. TPAMI 2014 , volume =

    2014

  65. [65]

    Razenshteyn and Ludwig Schmidt , title =

    Alexandr Andoni and Piotr Indyk and Thijs Laarhoven and Ilya P. Razenshteyn and Ludwig Schmidt , title =. Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015 , pages =

    2015

  66. [66]

    Jure Leskovec and Rok Sosic , title =

  67. [67]

    Herve Jegou and Romain Tavenard and Matthijs Douze and Laurent Amsaleg , title =

  68. [68]

    ICMR 2015 , pages =

    Shoou. ICMR 2015 , pages =

    2015

  69. [69]

    Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015 , pages =

    Xiang Zhang and Junbo Jake Zhao and Yann LeCun , title =. Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015 , pages =

    2015

  70. [70]

    Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics , pages =

    Tomas Mikolov and Wen. Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics , pages =

  71. [71]

    Accessed: 05-20-2019 , title =

    2019

  72. [72]

    2018 , Xnote =

    Microsoft AI & Research , month =. 2018 , Xnote =

    2018

  73. [73]

    VLDB'19 , year =

    Fu, Cong and Xiang, Chao and Wang, Changxu and Cai, Deng , title =. VLDB'19 , year =

  74. [74]

    FAQ: All about the Google RankBrain algorithm , author=

  75. [75]

    CoRR , volume =

    Lei Yu and Karl Moritz Hermann and Phil Blunsom and Stephen Pulman , title =. CoRR , volume =. 2014 , Xurl =

    2014

  76. [76]

    CIKM '16 , year =

    Van Gysel, Christophe and de Rijke, Maarten and Kanoulas, Evangelos , title =. CIKM '16 , year =

  77. [77]

    Bruce Croft , title =

    Mostafa Dehghani and Hamed Zamani and Aliaksei Severyn and Jaap Kamps and W. Bruce Croft , title =. SIGIR 2017 , pages =

    2017

  78. [78]

    Bruce Croft , title =

    Jiafeng Guo and Yixing Fan and Qingyao Ai and W. Bruce Croft , title =

  79. [79]

    Bhaskar Mitra and Fernando Diaz and Nick Craswell , title =

  80. [80]

    SIGIR 2017 , pages =

    Chenyan Xiong and Zhuyun Dai and Jamie Callan and Zhiyuan Liu and Russell Power , title =. SIGIR 2017 , pages =

    2017

Showing first 80 references.