CVE-2021-37679

HIGHNVD 7.17.1—
EchelonGraph scoreMEDIUM confidence

This high-severity CVE scores 7.1 under NVD CVSS v3. EPSS exploit probability: 0.2%, top 92% of all CVEs by exploit prediction. GitHub Security Advisory data not yet ingested — confidence will rise once GHSA publishes (typical lag: hours to days for open-source ecosystem CVEs; never for infrastructure-only CVEs).

Triggered by: NVD CVSS baseline
Sources: epss, nvd
Trending — 3 sources updated this week
7.1
EchelonGraph verdictPlan a fixSerious severity, but no confirmed exploitation yet.
  • High severity, but no confirmed exploitation yet
CISA-KEV: Not listedEPSS: 0%CVSS: 7.1Exploit: NoneExposed: 0

No vendor fix yet — apply a workaround or compensating control (WAF / firewall / segmentation) and watch for a patch.

TensorFlow is an end-to-end open source platform for machine learning. In affected versions it is possible to nest a tf.map_fn within another tf.map_fn call. However, if the input tensor is a RaggedTensor and there is no function signature provided, code assumes the output is a fully specified tensor and fills output buffer with uninitialized contents from the heap. The t and z outputs should be identical, however this is not the case. The last row of t contains data from the heap which can be used to leak other memory information. The bug lies in the conversion from a Variant tensor to a RaggedTensor. The implementation does not check that all inner shapes match and this results in the additional dimensions. The same implementation can result in data loss, if input tensor is tweaked. We have patched the issue in GitHub commit 4e2565483d0ffcadc719bd44893fb7f609bb5f12. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

CVSS v3
7.1
EG Score
7.1(medium)
EPSS
8.0%
KEV
Not listed

Published

August 12, 2021

Last Modified

November 21, 2024

References (4)

Affected Packages

(3 across 1 ecosystem)
PyPI(3)
PackageVulnerable rangeFixed inDependents
tensorflow2.3.0 ... 2.4.2 (7 versions)2.4.3—
tensorflow-cpu2.3.0 ... 2.4.2 (7 versions)2.4.3—
tensorflow-gpu2.3.0 ... 2.4.2 (7 versions)2.4.3—

Weakness Classification(2)

MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.

Data Freshness Timeline

(refreshed 14× in last 7d / 45× in last 30d)

Each row is a source pipeline that fetched or updated this CVE on that date, with what changed. For example, "NVD update" means NVD published or revised its analysis for this CVE; "MITRE cvelistV5" means we ingested or refreshed it from the CNA feed. Most recent first.

  1. 2026-07-22 22:49 UTCEG score recompute
  2. 2026-07-22 14:06 UTCEPSS rescore
  3. 2026-07-22 14:06 UTCEPSS rescore
  4. 2026-07-21 15:22 UTCEPSS rescore
  5. 2026-07-21 15:22 UTCEPSS rescore
  6. 2026-07-20 17:05 UTCEPSS rescore
  7. 2026-07-19 14:29 UTCEPSS rescore
  8. 2026-07-19 14:29 UTCEPSS rescore
  9. 2026-07-19 02:27 UTCEPSS rescore
  10. 2026-07-18 14:29 UTCOSV refresh
  11. 2026-07-18 10:02 UTCEPSS rescore
  12. 2026-07-18 10:02 UTCEPSS rescore
  13. 2026-07-16 17:00 UTCEPSS rescore
  14. 2026-07-16 17:00 UTCEPSS rescore
  15. 2026-07-15 16:55 UTCEPSS rescore
  16. 2026-07-15 01:58 UTCEPSS rescore
  17. 2026-07-13 22:27 UTCEPSS rescore
  18. 2026-07-13 06:10 UTCEPSS rescore
  19. 2026-07-12 05:44 UTCEPSS rescore
  20. 2026-07-12 05:44 UTCEPSS rescore
  21. 2026-07-11 08:25 UTCEPSS rescore
  22. 2026-07-11 08:25 UTCEPSS rescore
  23. 2026-07-09 19:07 UTCEPSS rescore
  24. 2026-07-08 15:12 UTCEPSS rescore
  25. 2026-07-07 13:43 UTCEPSS rescore
Show 75 more
  1. 2026-07-06 16:25 UTCEPSS rescore
  2. 2026-07-05 02:28 UTCEPSS rescore
  3. 2026-07-05 02:28 UTCEPSS rescore
  4. 2026-07-04 06:29 UTCEPSS rescore
  5. 2026-07-04 06:29 UTCEPSS rescore
  6. 2026-07-01 15:04 UTCEPSS rescore
  7. 2026-06-30 23:20 UTCEPSS rescore
  8. 2026-06-30 23:20 UTCEPSS rescore
  9. 2026-06-30 19:00 UTCOSV refresh
  10. 2026-06-29 14:04 UTCEPSS rescore
  11. 2026-06-28 14:05 UTCEPSS rescore
  12. 2026-06-28 04:54 UTCEPSS rescore
  13. 2026-06-28 04:54 UTCEPSS rescore
  14. 2026-06-27 03:06 UTCEPSS rescore
  15. 2026-06-25 13:48 UTCEPSS rescore
  16. 2026-06-25 13:48 UTCEPSS rescore
  17. 2026-06-24 14:03 UTCEPSS rescore
  18. 2026-06-24 14:03 UTCEPSS rescore
  19. 2026-06-23 21:31 UTCEPSS rescore
  20. 2026-06-23 21:31 UTCEPSS rescore
  21. 2026-06-22 14:24 UTCEPSS rescore
  22. 2026-06-22 14:24 UTCEPSS rescore
  23. 2026-06-21 14:55 UTCEPSS rescore
  24. 2026-06-21 01:58 UTCEPSS rescore
  25. 2026-06-21 01:58 UTCEPSS rescore
  26. 2026-06-19 19:24 UTCEPSS rescore
  27. 2026-06-19 19:24 UTCEPSS rescore
  28. 2026-06-18 17:51 UTCEPSS rescore
  29. 2026-06-18 17:51 UTCEPSS rescore
  30. 2026-06-17 17:51 UTCEPSS rescore
  31. 2026-06-16 17:51 UTCEPSS rescore
  32. 2026-06-16 17:51 UTCEPSS rescore
  33. 2026-06-15 17:46 UTCEPSS rescore
  34. 2026-06-14 23:15 UTCEPSS rescore
  35. 2026-06-13 22:58 UTCEPSS rescore
  36. 2026-06-13 22:58 UTCEPSS rescore
  37. 2026-06-13 00:14 UTCOSV refresh
  38. 2026-06-12 23:10 UTCEPSS rescore
  39. 2026-06-12 23:10 UTCEPSS rescore
  40. 2026-06-11 13:58 UTCEPSS rescore
  41. 2026-06-10 22:16 UTCEPSS rescore
  42. 2026-06-10 13:20 UTCEPSS rescore
  43. 2026-06-08 14:15 UTCEPSS rescore
  44. 2026-06-08 14:15 UTCEPSS rescore
  45. 2026-06-07 15:23 UTCEPSS rescore
  46. 2026-06-07 15:23 UTCEPSS rescore
  47. 2026-06-06 13:46 UTCEPSS rescore
  48. 2026-06-06 13:46 UTCEPSS rescore
  49. 2026-06-06 13:46 UTCEPSS rescore
  50. 2026-06-05 22:45 UTCEPSS rescore
  51. 2026-06-05 06:09 UTCEPSS rescore
  52. 2026-06-05 06:09 UTCEPSS rescore
  53. 2026-06-04 13:10 UTCEPSS rescore
  54. 2026-06-04 13:10 UTCEPSS rescore
  55. 2026-06-02 20:11 UTCEPSS rescore
  56. 2026-06-02 20:11 UTCEPSS rescore
  57. 2026-06-01 13:50 UTCEPSS rescore
  58. 2026-06-01 13:50 UTCEPSS rescore
  59. 2026-05-31 22:29 UTCEPSS rescore
  60. 2026-05-31 22:29 UTCEPSS rescore
  61. 2026-05-31 00:15 UTCEPSS rescore
  62. 2026-05-31 00:15 UTCEPSS rescore
  63. 2026-05-29 13:42 UTCEPSS rescore
  64. 2026-05-29 13:42 UTCEPSS rescore
  65. 2026-05-28 13:43 UTCEPSS rescore
  66. 2026-05-28 13:43 UTCEPSS rescore
  67. 2026-05-27 13:39 UTCEPSS rescore
  68. 2026-05-27 13:39 UTCEPSS rescore
  69. 2026-05-27 13:39 UTCEPSS rescore
  70. 2026-05-26 13:42 UTCEPSS rescore
  71. 2026-05-26 12:05 UTCEG score recompute
  72. 2026-05-26 07:17 UTCEPSS rescore
  73. 2026-05-26 07:17 UTCEPSS rescore
  74. 2026-05-26 07:17 UTCEPSS rescore
  75. 2026-05-26 02:16 UTCOSV refresh

Related CVEs(same CWE)

Frequently asked(5)

What is CVE-2021-37679?
CVE-2021-37679 is a high vulnerability published on August 12, 2021. TensorFlow is an end-to-end open source platform for machine learning. In affected versions it is possible to nest a tf.mapfn within another tf.mapfn call. However, if the input tensor is a RaggedTensor and there is no function signature provided, code assumes the output is a fully specified tensor…
When was CVE-2021-37679 disclosed?
CVE-2021-37679 was first published in the National Vulnerability Database on August 12, 2021, with the most recent update on November 21, 2024. EchelonGraph re-ingests CVE updates from NVD on a 2-hour cycle, so this page reflects the latest published state.
Is CVE-2021-37679 actively exploited?
CVE-2021-37679 is not currently on CISA's Known Exploited Vulnerabilities catalog. FIRST EPSS estimates a 8.0% percentile likelihood of exploitation in the next 30 days — higher percentiles indicate greater predicted risk.
What is the CVSS score of CVE-2021-37679?
CVE-2021-37679 has a CVSS v3 base score of 7.1 (NVD).
How do I remediate CVE-2021-37679?
Patch to the fixed version published by the affected vendor. Where vendor advisories exist for CVE-2021-37679, EchelonGraph cross-links them in the Vendor Advisories panel below — those typically contain the canonical remediation steps, fixed version numbers, and any vendor-specific mitigations.

Dependency Blast Radius

See which npm, PyPI, Go, and Maven packages are affected by CVE-2021-37679

Explore →

Is Your Infrastructure Affected by CVE-2021-37679?

EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.

Start Free Scan →Browse All CVEs