Benchmark Almanac

TalentCLEF 2025 Job Title-Based Skill Prediction

Alibaba-NLP/gte-Qwen2-7B-instruct 0.36 tops its leaderboard
Not in model cards
Version 2025

Models rank skills for an English job title. Organizers compare the ranking with annotated skills using mean average precision.

The official results include 17 runs. The leading run used gte-Qwen2-7B-instruct and scored 0.360 mean average precision.

Made by
Avature and UNED
Released
September 2024
Kind
Other professions
Level
University
Scored
Checked automatically
Answers
Short answer
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Run by
A host
9
Entries on its own leaderboard, 0 models we track

Alibaba-NLP/gte-Qwen2-7B-instruct tops the leaderboard with 0.36 No scores yet

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran on its own with several settings, we keep its best. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. Agents built around a model, such as HULAT-UC3M. + Gemini 2.5 Flash, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best.

  1. 1 Alibaba-NLP/gte-Qwen2-7B-instruct 278954, Retrieval, Contrastive learning, CachedGIST, Data Augmentation 0.36
  2. 2 Multilingual E5 Large Instruct 284333, Retrieval, Contrastive learning, InfoNCE, Data Augmentation, ESCO data 0.34
  3. 3 BAAI/bge-large-en-v1.5 278957, Retrieval, Contrastive learning, GIST, Data Augmentation 0.34
  4. 4 microsoft/mdeberta-v3-base 279245, Retrieval, Classification, CrossEntropy, No usage 0.29
  5. 5 all-MiniLM-L6v2 284949, Retrieval, Ensemble, Reranking 0.28
  6. 6 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 284992, Retrieval, Contrastive learning, InfoNCE, No usage, Internal data 0.25
  7. 7 paraphrase-multilingual-mpnet-base-v2 284784, Retrieval, Zero-shot, No usage 0.22
  8. 8 snowflake-arctic-embed-l-v2.0 280606, Retrieval+Reranking, Zero-shot, Reranking 0.14
  9. 9 HULAT-UC3M. + Gemini 2.5 Flash Sep 2025, 282011, Reranking, Zero-shot, Reranking 0.11

Source: Official TalentCLEF 2025 Job Title-Based Skill Prediction leaderboard

#EntryScoreSettingSubmitted byDateSource
1 Alibaba-NLP/gte-Qwen2-7B-instruct 7B 0.36 278954, Retrieval, Contrastive learning, CachedGIST, Data Augmentation, pjmathematician pjmathematician Official leaderboard
2 Multilingual E5 Large Instruct 0.6B 0.34 284333, Retrieval, Contrastive learning, InfoNCE, Data Augmentation, ESCO data, moali moali Official leaderboard
3 BAAI/bge-large-en-v1.5 0.3B 0.34 278957, Retrieval, Contrastive learning, GIST, Data Augmentation, pjmathematician pjmathematician Official leaderboard
4 microsoft/mdeberta-v3-base 0.3B 0.29 279245, Retrieval, Classification, CrossEntropy, No usage, NLPnorth NLPnorth Official leaderboard
5 all-MiniLM-L6v2 4B 0.28 284949, Retrieval, Ensemble, Reranking, iagox iagox Official leaderboard
6 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 0.3B 0.25 284992, Retrieval, Contrastive learning, InfoNCE, No usage, Internal data, Techwolf Techwolf Official leaderboard
7 paraphrase-multilingual-mpnet-base-v2 0.3B 0.22 284784, Retrieval, Zero-shot, No usage, SkillSeekers SkillSeekers Official leaderboard
8 snowflake-arctic-embed-l-v2.0 0.6B 0.14 280606, Retrieval+Reranking, Zero-shot, Reranking, HULAT-UC3M. HULAT-UC3M. Official leaderboard
9 HULAT-UC3M. + Gemini 2.5 Flash Sep 2025 0.11 282011, Reranking, Zero-shot, Reranking HULAT-UC3M. Sep 2025 Official leaderboard
#ModelScoreSettingSubmitted byDateSource

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

Versions

More in Law, finance and medicine

All 637 in Law, finance and medicine
  • MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy. Qwen 3.8 Max 80.4%
  • HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
  • HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy. Ling-3.1-flash 81.7%
  • KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
  • Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (9 entries).
  • Pass Models scored 9 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on talentclef.github.io: 1 row for models we track and 8 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check