Up tókànItọsọna atẹle
Iṣiro-ara-ẹni ni Awọn Yipo Aṣoju
Imọ-ẹrọ
Awọn ohun elo Itọsọna
Reflexion ati Awọn aṣoju Atunse Ara-ẹni
Reflexion is a research framework in which language agents use task feedback to write textual reflections for later attempts.
Gbọ si itọsọna yiiAI-ṣe afihan · ~1 min
Lori iwe yi1 iseju kika
Akopọ
It changes context, not model weights.
Jin Dive
The Reflexion paper describes an Actor that attempts a task, an Evaluator that scores the result, and a Self-Reflection model that turns feedback into language stored in episodic memory. The next attempt can use that reflection as context; the framework does not update model weights. The paper reports improvements on selected tasks, including a HumanEval result for its particular setup. Benchmark results should not be generalized to every agent or deployment. In the Reflexion paper’s reported coding experiment, the framework achieved 91% pass@1 on HumanEval, compared with 80% for the GPT-4 baseline described in that paper. These are results for the paper’s particular setup and benchmark, not a general performance guarantee.
Real-World imuse
An agent receives unit-test feedback, writes a reflection, and retries a coding task.
A researcher compares Reflexion’s benchmark setup with its stated baseline.
An evaluator supplies task-specific feedback that the agent can inspect.
A developer stops iteration when a defined success condition is met.
Ojo iwaju ti ifasilẹ ati Awọn aṣoju Atunse Ara-ẹni
Self-correction methods remain dependent on evaluation quality, task design, and iteration cost. Avoid claims of general improvement without testing; reflections can be mistaken or lead to loops.
Fifi o sinu iwa
Kini idi ti o ṣe pataki
- Kọ awọn yiyan. Apẹrẹ ipele-ohun elo pinnu boya AI ṣe ilọsiwaju awọn abajade gidi.
- Ẹgbẹ ati ṣiṣan iṣẹ. Ijọpọ iṣan-iṣẹ ti o dara ṣẹda awọn anfani iṣẹ-ṣiṣe ti awọn olumulo le gbẹkẹle.
- Ewu ati ailewu. Awọn ọran lilo ti iwọn daradara dinku rirẹ iyipada ati eewu imuse.
Ṣọra fun
- Ṣiṣẹda ilana fifọ le ṣe alekun awọn iṣoro to wa tẹlẹ.
- Awọn ẹgbẹ le ṣe adaṣe adaṣe ki o yọ idajọ eniyan ti o nilo kuro.
- Didara le fò ti awọn abajade ko ba ni iṣiro nigbagbogbo.
Next awọn igbesẹ
- Ṣe maapu iṣan-iṣẹ lọwọlọwọ ki o ṣe idanimọ igbesẹ ti o ga julọ.
- Ṣe alaye awọn aaye ayẹwo eniyan ṣaaju adaṣe ni kikun.
- Kọ awọn olumulo lori awọn itọsi, awọn ọna igbega, ati awọn iṣedede didara.
- Tọpinpin awọn abajade ipele-ṣiṣe lati jẹrisi iye idaduro.
Tesiwaju Ṣiṣawari
Ìwé ìròyìn ọfẹ
Gba ìkànìyàn AI ojoojúmọ́
Ìtàn AI mẹta tó jẹ́rìí ní gbogbo owurọ́ ọjọ́ iṣẹ́, kọ́wé ní Gẹ̀ẹ́sì tó rọrùn. Ọfẹ́ fún gbogbo àkókò, kò sí ìpolówó.
Ìmeèlì kan ní gbogbo ọjọ́ iṣẹ́. Yọkúrò ní ìkànìyàn ní ìkànìyàn kan. A kò ta tàbí pín àdírẹ́sì rẹ.
Ṣe idanwo fun ara rẹ
Take the Reflexion and Self-Correcting Agents quiz
Ìdáhùn lẹ́sẹ̀kẹsẹ̀ lórí gbogbo ìdáhùn, àti iwe-ẹ̀rí tó lè pin pẹ̀lú ID tó lè jẹ́rìí lẹ́yìn tí o bá kọ́ ẹ̀kọ́.
Ṣe àtìlẹ́yìn ẹ̀kọ́ AI ọfẹ́. AI Understanding jẹ́ àjọ àìní èrè 501(c)(3) — kò sí ìpolówó, kò sí àkúnya owó, ní gbogbo àkókò. Ṣe ẹbun kan
Awọn ibeere ti a beere nigbagbogbo
What is Reflexion and Self-Correcting Agents?
Reflexion is a research framework in which language agents use task feedback to write textual reflections for later attempts. It changes context, not model weights.
Kini abuda asọye ti bii aṣoju Reflexion kan ṣe 'kọ ẹkọ'?
Iṣatunṣe nigba miiran ni a pe ni 'ẹkọ imuduro ọrọ ẹnu' nitori awọn ẹkọ ti wa ni ipamọ bi ọrọ ede adayeba ni iranti, kii ṣe koodu sinu awọn aye awoṣe.
Ninu ilana Isọdọtun, kini Oluyẹwo ṣe?
Oluyẹwo (idanwo ẹyọkan kan, ṣayẹwo-baramu deede, tabi adajọ LLM) ṣe agbejade ifihan agbara ti igbesẹ Irohin-ara-ẹni yipada si ẹkọ ọrọ.
Kilode ti didara ifihan agbara igbelewọn ṣe pataki pupọ ninu Ipadabọ?
Ti oluyẹwo ko ba ni igbẹkẹle, aṣoju le kọ awọn ifarabalẹ ti o ni igboya ti o da lori awọn esi buburu, ṣiṣe awọn igbiyanju iwaju ni ọna ti ko tọ.
On which coding benchmark did the Reflexion paper report a 91% pass@1 result?
The paper reports 91% pass@1 on HumanEval for its Reflexion coding setup; the result is specific to that evaluation.
Ewo ni eewu ṣiṣi otitọ pẹlu awọn aṣoju ti n ṣatunṣe ara ẹni?
Laisi ofin idaduro to dara, aṣoju le tẹsiwaju atunṣe awọn idahun ti o pe, jafara iṣiro ati nigbakan ti o bajẹ abajade to dara.
Tesiwaju kikọ
Jẹmọ awọn itọsọna
Awọn itọsọna diẹ sii ti a yan fun koko yii