# Discriminative Reasoning on RPEval Multi-MICRO Explicit Memory

Source: [Discriminative Reasoning on RPEval Multi-MICRO Explicit Memory](<https://www.sota2.com/research/sota/discriminative-reasoning-on-rpeval-multi-micro-explicit-memory>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:46:46.972Z

Evaluation of model performance on the RPEval Multi-MICRO Explicit Memory dataset for the Discriminative Reasoning task, measured by IA, LKN, and ALL scores.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Discriminative Reasoning |
| Dataset (including any stated split) | RPEval Multi-MICRO Explicit Memory |
| Metric | IA Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [IA Score](<https://www.sota2.com/research/sota/discriminative-reasoning-on-rpeval-multi-micro-explicit-memory.md?metric=IA%20Score>): Higher is better; unit: SCALAR.

- [LKN Score](<https://www.sota2.com/research/sota/discriminative-reasoning-on-rpeval-multi-micro-explicit-memory.md?metric=LKN%20Score>): Higher is better; unit: SCALAR.

- [ALL Score](<https://www.sota2.com/research/sota/discriminative-reasoning-on-rpeval-multi-micro-explicit-memory.md?metric=ALL%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | IA Score | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-5 + RP-Reasoner | 0.71 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-4.1 + RP-Reasoner | 0.69 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| DeepSeek-V3 + CoT | 0.67 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| DeepSeek-V3 + RP-Reasoner | 0.67 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| DeepSeek-V3 + Reminder | 0.57 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-4.1 + CoT | 0.56 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| Qwen2.5-7B + RP-Reasoner | 0.55 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| DeepSeek-V3 | 0.55 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-4.1 + Reminder | 0.52 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| Qwen2.5-7B | 0.48 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-4.1 | 0.48 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| Qwen2.5-7B + Reminder | 0.45 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-5 + CoT | 0.38 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-5 | 0.31 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| GPT-5 + Reminder | 0.26 | Not recorded | 6976f4e6a2b28e761d87c0ca |
| Qwen2.5-7B + CoT | 0.18 | Not recorded | 6976f4e6a2b28e761d87c0ca |

### GPT-5 + RP-Reasoner

IA Score: 0.71. Result ID: 699671c05698665ec4b8e7de.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"GPT-5","Prompt Strategy":"RP-Reasoner"}. This record does not assert independent reproduction.

### GPT-4.1 + RP-Reasoner

IA Score: 0.69. Result ID: 699671c05698665ec4b8e7da.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"GPT-4.1","Prompt Strategy":"RP-Reasoner"}. This record does not assert independent reproduction.

### DeepSeek-V3 + CoT

IA Score: 0.67. Result ID: 699671c05698665ec4b8e7d5.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77001e209131cea7066a.md>)

Recorded evaluation configuration: {"Base Model":"DeepSeek-V3","Prompt Strategy":"Chain-of-Thought"}. This record does not assert independent reproduction.

### DeepSeek-V3 + RP-Reasoner

IA Score: 0.67. Result ID: 699671c05698665ec4b8e7d6.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"DeepSeek-V3","Prompt Strategy":"RP-Reasoner"}. This record does not assert independent reproduction.

### DeepSeek-V3 + Reminder

IA Score: 0.57. Result ID: 699671c05698665ec4b8e7d4.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77001e209131cea7066a.md>)

Recorded evaluation configuration: {"Base Model":"DeepSeek-V3","Prompt Strategy":"Reminder"}. This record does not assert independent reproduction.

### GPT-4.1 + CoT

IA Score: 0.56. Result ID: 699671c05698665ec4b8e7d9.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Base Model":"GPT-4.1","Prompt Strategy":"Chain-of-Thought"}. This record does not assert independent reproduction.

### Qwen2.5-7B + RP-Reasoner

IA Score: 0.55. Result ID: 699671c05698665ec4b8e7d2.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Prompt Strategy":"RP-Reasoner"}. This record does not assert independent reproduction.

### DeepSeek-V3

IA Score: 0.55. Result ID: 699671c05698665ec4b8e7d3.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77001e209131cea7066a.md>)

Recorded evaluation configuration: {"Base Model":"DeepSeek-V3","Prompt Strategy":"Vanilla"}. This record does not assert independent reproduction.

### GPT-4.1 + Reminder

IA Score: 0.52. Result ID: 699671c05698665ec4b8e7d8.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Base Model":"GPT-4.1","Prompt Strategy":"Reminder"}. This record does not assert independent reproduction.

### Qwen2.5-7B

IA Score: 0.48. Result ID: 699671c05698665ec4b8e7cf.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Prompt Strategy":"Vanilla"}. This record does not assert independent reproduction.

### GPT-4.1

IA Score: 0.48. Result ID: 699671c05698665ec4b8e7d7.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Base Model":"GPT-4.1","Prompt Strategy":"Vanilla"}. This record does not assert independent reproduction.

### Qwen2.5-7B + Reminder

IA Score: 0.45. Result ID: 699671c05698665ec4b8e7d0.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Prompt Strategy":"Reminder"}. This record does not assert independent reproduction.

### GPT-5 + CoT

IA Score: 0.38. Result ID: 699671c05698665ec4b8e7dd.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"GPT-5","Prompt Strategy":"Chain-of-Thought"}. This record does not assert independent reproduction.

### GPT-5

IA Score: 0.31. Result ID: 699671c05698665ec4b8e7db.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"GPT-5","Prompt Strategy":"Vanilla"}. This record does not assert independent reproduction.

### GPT-5 + Reminder

IA Score: 0.26. Result ID: 699671c05698665ec4b8e7dc.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

Recorded evaluation configuration: {"Base Model":"GPT-5","Prompt Strategy":"Reminder"}. This record does not assert independent reproduction.

### Qwen2.5-7B + CoT

IA Score: 0.18. Result ID: 699671c05698665ec4b8e7d1.

[Reporting paper](<https://www.sota2.com/research/paper/6976f4e6a2b28e761d87c0ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Prompt Strategy":"Chain-of-Thought"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
