InayofuataMwongozo unaofuata
Kujitafakari katika Mizunguko ya Wakala
Kiufundi
MWONGOZO wa Maombi
Mawakala wa Kutafakari na Kujisahihisha
Reflexion is a research framework in which language agents use task feedback to write textual reflections for later attempts.
Sikiliza mwongozo huuInaonyeshwa na AI · ~1 min
Katika ukurasa huu1 dakika kusoma
Muhtasari
It changes context, not model weights.
Dive ya kina
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.
Utekelezaji wa Ulimwengu Halisi
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.
Mustakabali wa Mawakala wa Kutafakari na Kujisahihisha
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.
Kuiweka katika vitendo
Kwa nini ni muhimu
- Tengeneza chaguzi. Muundo wa kiwango cha programu huamua kama AI inaboresha matokeo halisi.
- Timu na mtiririko wa kazi. Ujumuishaji mzuri wa mtiririko wa kazi hutengeneza faida za tija ambazo watumiaji wanaweza kuamini.
- Hatari na usalama. Kesi za utumiaji zilizopangwa vizuri hupunguza uchovu wa mabadiliko na hatari ya utekelezaji.
Jihadharini na
- Kuweka kiotomatiki mchakato uliovunjika kunaweza kukuza shida zilizopo.
- Timu zinaweza kufanya otomatiki kupita kiasi na kuondoa uamuzi unaohitajika wa kibinadamu.
- Ubora unaweza kuyumba ikiwa matokeo hayatatathminiwa mara kwa mara.
Hatua zinazofuata
- Ramani ya mtiririko wa kazi wa sasa na utambue hatua ya msuguano wa juu zaidi.
- Bainisha vituo vya ukaguzi vya binadamu kabla ya otomatiki kamili.
- Fundisha watumiaji kuhusu maekelezo, njia za kupanda na viwango vya ubora.
- Fuatilia matokeo ya kiwango cha kazi ili kuthibitisha thamani endelevu.
Endelea Kuchunguza
Jarida halali
Pata taarifa ya kila siku ya AI
Hadithi tatu za AI zilizothibitishwa kila asubuhi ya wiki, zimetengenezwa kwa Kiingereza rahisi. Bila malipo daima, hakuna matangazo.
Barua pepe moja kila siku ya wiki. Jiondoe kwa bonyeza moja. Hatutauza au kushiriki anwani yako.
Jijaribu mwenyewe
Take the Reflexion and Self-Correcting Agents quiz
Maoni ya haraka juu ya kila jibu, na cheti kinachoweza kushirikishwa na kitambulisho kinachoweza kuthibitishwa baada ya kupita kozi.
Changia elimu ya AI bila malipo. AI Understanding ni 501(c)(3) isiyo na faida — hakuna matangazo, hakuna kizuizi cha malipo, daima. Toa mchango
Maswali yanayoulizwa mara kwa mara
Mawakala wa Kutafakari na Kujisahihisha ni nini?
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.
Ni nini sifa bainifu ya jinsi wakala wa Reflexion 'hujifunza'?
Reflexion wakati mwingine huitwa 'kujifunza kwa uimarishaji wa maneno' kwa sababu masomo huhifadhiwa kama maandishi ya lugha asili kwenye kumbukumbu, sio kusimba katika vigezo vya muundo.
Katika mfumo wa Reflexion, Mtathmini hufanya nini?
Mtathimini (jaribio la kitengo, hundi inayolingana kabisa, au mwamuzi wa LLM) hutoa ishara kwamba hatua ya Kujitafakari inageuka kuwa somo la maandishi.
Kwa nini ubora wa ishara ya tathmini ni muhimu sana katika Reflexion?
Ikiwa mtathmini hawezi kutegemewa, wakala anaweza kuandika tafakari za kujiamini kulingana na maoni mabaya, akielekeza majaribio ya siku zijazo katika mwelekeo usio sahihi.
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.
Ni hatari gani ya kweli iliyo wazi na mawakala wa kujisahihisha?
Bila sheria nzuri ya kusimamisha, wakala anaweza kuendelea kusahihisha majibu sahihi, kupoteza hesabu na wakati mwingine kudhalilisha matokeo mazuri.
Endelea kujifunza
Miongozo inayohusiana
Miongozo zaidi imechaguliwa kwa mada hii