Backtracing
Models identify the sentence that prompted a student's question or a news reader's curiosity. In conversations, they find the turn that prompted an emotional response.
The evaluation covers 210 lecture queries, 1,382 news queries, and 671 conversation queries. The authors report top-one and top-three accuracy by domain.
- Made by
- Rose E. Wang and colleagues
- Released
- March 2024
- Kind
- Finding facts in long inputs
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
gpt-3.5-turbo-16k tops the leaderboard with 43%
Best score per entry from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this is the plain mean of top-one accuracy across Lecture, News Article and Conversation. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 1 gpt-3.5-turbo-16k single-sentence 43%
- 2 OPT 6B single-sentence 32.7%
- 3 Re-ranker retrieval method 32%
- 4 GPTJ single-sentence 31.7%
- 5 Cross-Encoder retrieval method 29.7%
- 6 Bi-Encoder (all-MiniLM) retrieval method 25.3%
- 7 Bi-Encoder (Q&A) retrieval method 24%
- 8 GPT2 single-sentence 22%
- 9 Edit retrieval method 4.00%
Source: The Backtracing paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | gpt-3.5-turbo-16k | 43% | single-sentence | Official leaderboard | |
| 2 | OPT 6B | 32.7% | single-sentence | Official leaderboard | |
| 3 | Re-ranker | 32% | retrieval method | Official leaderboard | |
| 4 | GPTJ | 31.7% | single-sentence | Official leaderboard | |
| 5 | Cross-Encoder | 29.7% | retrieval method | Official leaderboard | |
| 6 | Bi-Encoder (all-MiniLM) | 25.3% | retrieval method | Official leaderboard | |
| 7 | Bi-Encoder (Q&A) | 24% | retrieval method | Official leaderboard | |
| 8 | GPT2 | 22% | single-sentence | Official leaderboard | |
| 9 | Edit | 4.00% | retrieval method | Official leaderboard |
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.
More in Long context
- MRCR v2 In use Multi-round coreference: retrieve the n-th of several similar requests hidden in a long synthetic conversation, up to 1M tokens.
Muse Spark 1.3 98.5%
- LongBench v2 In use 503 multiple-choice questions over 8K-2M-word contexts (QA, dialogue, code repos, structured data). Accuracy.
Inkling 2 Small 70.6%
- GraphWalks In use A graph is given as an edge list in long context. The model answers BFS or parent queries, scored by F1. Claude Fable 5 99.96%
- OpenAI MRCR v2 In use Models read a long synthetic chat that hides 2, 4 or 8 identical writing requests and must reproduce the response to the i-th one. Scored by mean string match ratio per context bin.
Gemini 3.7 Flash 89%
- CL-bench In use Model must learn new knowledge from a supplied long context (rules, procedures, data) and solve tasks. Strict LLM-verified rubrics. GPT-5.4 27.9%
- AA-LCR 1.1 In use Models answer 100 questions that need reasoning across sets of real documents of about 100K tokens. An LLM equality checker with a fixed system prompt compares each answer with the key.
Step 5 Preview 88.3%
Data audit
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (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 No scores yet.
Leaderboard import 3 pass
- Pass Importer We read the board on aclanthology.org: 9 entries at the last read, none a model we track.
- Pass Last read November 21st, 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.