閱讀理解之科普說明文-2025年新高考英語模擬試題分類匯編（解析版）

2025年新高考最新英語模擬試題優(yōu)選分類匯編

閱讀理解之科普說明文

森贏i

科普說明文是高考的重要話題，也是國家的重要導向。其在英語高考試題中的地位也就不|

言而喻。本資料選取了全國模擬試題中的科普說明文素材進行了匯編，以滿足廣大師生的需求」

1.（2025?廣東省深圳而羅湖區(qū)高三上季第一次摸底）Formanyyearsresearchersfocusedonwhatpeopleknow

aboutscience,thinkingthat“Toknowscienceistoloveit."Butdopeoplewhothinktheyknowscienceactuallyknow

science?AnewstudyledbyCristinaFonsecaoftheGeneticsSociety,LaurenceHurstoftheMilnerCentreforEvolution

（進化）revealsthatpeoplewithstrongattitudestendtobelievetheyunderstandscience,whileneutrals（中立者）areless

certain.Absoluteattitudes,bothforandagainst,buildonhighself-confidenceinknowledgeaboutscience.

Thestudyperformedasurveyofover2,000UKadults,askingthembothabouttheirattitudestoscienceandtheir

beliefintheirownunderstanding.Questionsfocusedongenetic（基因的）science,forexample,66Howwouldyourateyour

understandingofwhatthetermDNAmeans?^^Allindividualswerescoredfromzero（theyknowtheyhaveno

understanding）toone（theyareconfidentthattheyunderstand）.Theresultssuggestthatthoseattheattitudinalextremes——

bothstronglysupportiveandfirmlyopposing-haveveryhighself-beliefintheirowncomprehension,whilethose

answeringneutrallydonot.

Psychologically,theteamsuggests,thismakessense:toholdastrongopinionyouneedtostronglybelieveinthe

correctnessofyourinterpretationofthebasicfacts.Resultsofpreviousstudiesalsoindicatethatthosemoreacceptingof

sciencebothbelievetheyunderstanditandscorewellonthetextbookfact（true/false）questions.Bycontrast,peoplewith

strongnegativeattitudestosciencetendtobeoverconfidentabouttheirlevelofunderstanding.

WhetheritbeclimatechangeorGMfoods,importantsciencecaninspirestrongandopposingattitudes.

Understandinghowtocommunicatesciencerequiresanawarenessofwhypeoplemayholdsuchdifferentattitudestoit.

Whenitwasthoughtthatwhatmatteredmostforscientificliteracywasscientificknowledge,sciencecommunication

focusedonpassinginformationfromscientiststothepublic.However,thisapproachmaynotbesuccessful,andinsome

casescanhaveadverseeffects.Workingtoaddressthegapbetweenwhatpeopleknowandwhattheybelievetheyknow

maybeabetterstrategy.

8.Whatisthenewstudymainlyabout?

A.Anassessmentofpeople'sinterestingenetics.

B.Asurveyofvariousattitudestowardsevolution.

C.Areportofpeople'sgeneralknowledgeofscience.

D.Ananalysisoffactorsonpeople'sbeliefinscience.

9.Whydoestheauthormentionpreviousstudiesinparagraph3?

A.Tosupportthefindingsofthestudy.

B.Toclarifytheconceptofconfidence.

C.Tostresstheimportanceofbasicfacts.

D.Tocomparedifferentresearchmethods.

10.Whatdoestheunderlinedword“adverse“meaninthelastparagraph?

A.Random.

B.Negative.

C.Indirect.

D.Favorable.

11.Asforsciencecommunication,whichwouldtheauthoragreewith?

A.Itshouldhelppeopleboostconfidence.

B.Itshouldfocusonscientificknowledge.

C.Itshouldinspirepeopletoscorewellintests.

D.Itshouldemphasizeobjectiveawarenessofoneself.

【答案】8.D9.A10.B11.D

【解析】本文為一篇說明文，本文圍繞人們對于科學信仰展開，一項研究表明，態(tài)度強硬的人傾向于認為自己了解

科學，而態(tài)度中立的人對此不太確定。絕對的態(tài)度建立在對科學知識的高度自信之上。

8.細節(jié)理解題。根據文章第二段“Thestudyperformedasurveyofover2,000UKadults,askingthembothabouttheir

attitudestoscienceandtheirbeliefintheirownunderstanding.（這項研究對2000多名英國成年人進行了調查，詢問他們

對科學的態(tài)度和對自己理解的信念）”可知，該研究主要以影響人們科學信仰的因素分析展開。故選D。

9.推理判斷題。根據文章第三段“Psychologically,theteamsuggests,thismakessense:toholdastrongopinionyouneed

tostronglybelieveinthecorrectnessofyourinterpretationofthebasicfacts.Resultsofpreviousstudiesalsoindicatethat

thosemoreacceptingofsciencebothbelievetheyunderstanditandscorewellonthetextbookfact（true/false）

questions.（該團隊認為，從心理學上講，這是有道理的：要持有強烈的觀點，你需要堅信自己對基本事實的解釋是

正確的。之前的研究結果也表明，那些更接受科學的人認為他們理解它，并在教科書的事實（正確/錯誤）問題上獲

得高分）”可知，作者提及之前的研究是為了支持這項研究的結果，即，那些態(tài)度極端的人——包括強烈支持和堅決

反對一對自己的理解有很高的自信。故選A。

10.詞義猜測題。根據文章最后一段“However,thisapproachmaynotbesuccessfill（然而，這種方法可能不會成功）”

及“andinsomecasescanhave（在一些情況下可能有）”及“effects（影響）”可知，這種方法不會成功，甚至在一些情況下

會有更消極的影響，劃線詞adverse和negative意思相近。故選B。

11.推理判斷題。根據文章最后一段"Workingtoaddressthegapbetweenwhatpeopleknowandwhattheybelievethey

knowmaybeabetterstrategy.（努力解決人們知道的和他們認為他們知道的之間的差距可能是一個更好的策略）”可知，

作者認為在科學傳播上，應努力解決人們主觀上和客觀的認知差距，即，應強調客觀的自我意識。故選D。

2.（2025?河北省衡水中學第一次綜合素養(yǎng)測評）Evolution（進化）canperformextraordinarymakeovers;today's

airbornesongbirdsevolvedfromthewingless,earthbounddinosaursthatwanderedmillionsofyearsago.Butsome

organismsseemtobeunchanged-inotherwords,escapenaturalselection.Thecoelacanth,amodern-dayfish,isnearly

identicaltoits41O-million-year-oldfossils.

Scientistshavelongwonderedhowthesespeciesdoso.Ithasbeenassumedthatnaturalselectionkeepssomespecies

unchangedbyselectingformoderateoraveragequalities（stabilizingselection）ratherthanselectingformoreextreme

qualitiesthatwouldcauseaspeciestochange（directionalselection）.

ButastudypublishedintheNationalAcademyofSciencesUSAcontradictsthisidea,showingthatevolution

constantlyfavorsdifferentqualitiesinseeminglyunchanginganimalstoimproveshort-termsurvival.Inthelongterm,

though,t6allthatevolutioncancelsoutandleadstonochange,9,saysthestudy'sleadauthor,JamesStroud.

Stroudandhiscolleaguesstudiedforlizard（蜥蜴）specios;allrelativelyunchangedfor20millionyears.The

researcherscaughtmembersofthesepopulationseverysixmonthsforthreeyears.Theymeasuredeachlizard'sheadsize,

leglength,massandheight,aswellasthesizeofitsstickytoes（腳趾頭）,notingwhichindividualssurvived.Stroud

expectedtoobservestabilizingselectionatworkpreservingmoderatequalities.Insteadhesawclearerevidenceof

directionalselection:somelizardswithuniquecharacteristics,suchasstickiertoes,survivedbetter.

“Thestudyoffersagoodexplanationforwhyweseewhatwethinkisstabiliringselection,9,saysTadashiFukami,an

ecologiststudyingevolutionatStanfordUniversity.Manynewqualitiesareevolvingintheshortterm,buttheydon't

provideacrucialadvantageoverthelongterm.Inotherwords,speciesstayingunchangedmaysimplyhavefoundthebest

possiblecombinationofqualitiesforlastingsuccessintheirenvironment.Sowhathappenswhenthelizards5environment

changesmoredramatically?Tohelpanswerthisbiggerquestion,Stroudisstillmakingtripstovisitthelizards.

8.Whydoestheauthormentionthe<4coelacanthfish“inparagraph1?

A.Todemonstratethepowerofevolution.B.Toaddevidencetonaturalselection.

C.Togiveanexampleofunchangedspecies.D.Toprovespecies*extraordinarymakeovers.

9.What'sthefocusofparagraph2?

A.Unsolvedmysteries.B.Acommonbelief.

C.Asharpcontrast.D.UniqueCharacteristics.

10.HowdidStroudconducthisresearch?

A.Byanalyzinglizardfossils.B.Bytrackingresearchobjects.

C.Byillustratingstabilizingselection.D.Bycategorizingqualitiesoflizards.

11.WhatwillStroudprobablydointhefuture?

A.Maketripstovisitlizardexperts.B.Summarizeaveragefeaturesoflizards.

C.Revealthebestcombinationsofqualities.D.Examinelizardsunderextremeconditions.

【答案】8.C9.B10.B11.D

【解析】本文是說明文。文章通過介紹科學家對看似沒有發(fā)生變化的生物物種進行研究，揭示了這些物種在進化過

程中的選擇機制。

8.推理判斷題。根據第一段“Evolution（進化）canperformextraordinarymakeovers;today'sairbornesongbirdsevolved

fromthewingless,earthbounddinosaursthatwanderedmillionsofyearsago.Butsomeorganismsseemtobeunchanged

一inotherwords,escapenaturalselection.Thecoelacanth,amodern-dayfish,isnearlyidenticaltoits41O-million-year-old

fossils.（進化能夠產生非凡的變化；今天在空中飛翔的鳴鳥是從數(shù)百萬年前無翅、陸生的恐龍演化而來的。但有些生

物似乎沒有改變——換句話說，它們逃脫了自然選擇?，F(xiàn)代魚類腔棘魚與其4.1億年前的化石幾乎一模一樣）”可知，

本段提到腔棘魚是為了舉一個物種不變的例子。故選C項。

9.主旨大意題。根據第二段“Scientistshavelongwonderedhowthesespeciesdoso.Ithasbeenassumedthatnatural

selectionkeepssomespeciesunchangedbyselectingformoderateoraveragequalities（stabilizingselection）ratherthan

selectingformoreextremequalitiesthatwouldcauseaspeciestochange（directionalselection）」科學家們長期以來一直

想知道這些物種是如何做到這一點的。人們一直認為，自然選擇通過選擇適中或平均的品質（穩(wěn)定選擇）來保持某

些物種的不變性，而不是選擇會導致物種發(fā)生變化的更極端的品質（定向選擇）。）”可知，本段主要講述了人們一

直認為自然選擇是穩(wěn)定選擇而不是定向選擇，也就是一個共同的信念。故選B項。

10.推理判斷題。根據倒數(shù)第二段“Stroudandhiscolleaguesstudiedforlizard（蜥蜴）specios;allrelativelyunchangedfor

20millionyears.Theresearcherscaughtmembersofthesepopulationseverysixmonthsforthreeyears.Theymeasured

eachlizard'sheadsize,leglength,massandheight,aswellasthesizeofitsstickytoes（腳趾頭）,notingwhichindividuals

survived.Stroudexpectedtoobservestabilizingselectionatworkpreservingmoderatequalities.Insteadhesawclearer

evidenceofdirectionalselection:somelizardswithuniquecharacteristics,suchasstickiertoes,survivedbetter.（斯特勞德

和他的同事研究了蜥蜴物種；所有這些在2000萬年間都相對沒有變化。研究人員連續(xù)三年每六個月捕捉這些種群

中的成員。他們測量了每只蜥蜴的頭部大小、腿長、體重和身高，以及粘性腳趾的大小，并記錄了哪些個體存活了

下來。斯特勞德原本希望觀察到穩(wěn)定選擇在保持中等品質方面的作用o然而，他看到的卻是更明確的定向選擇證據：

一些具有獨特特征的蜥蜴，比如更粘的腳趾，存活率更高）”可知，斯特勞德是通過跟蹤研究對象來進行他的研究的。

故選B項。

11.推理判斷題。根據最后一段"Inotherwords,speciesstayingunchangedmaysimplyhavefoundthebestpossible

combinationofqualitiesforlastingsuccessintheirenvironment.Sowhathappenswhenthelizards'environmentchanges

moredramatically?Tohelpanswerthisbiggerquestion,Stroudisstillmakingtripstovisitthelizards.（換句話說，保持不變

的物種可能只是找到了在其環(huán)境中持久成功的最佳品質組合。那么，當蜥蜴的環(huán)境發(fā)生更劇烈的變化時會發(fā)生什么

呢？為了幫助回答這個更大的問題，斯特勞德仍在繼續(xù)探訪這些蜥蜴）”可知，接下來斯特勞德會在極端條件下檢查

蜥蜴。故選D項。

3.（2025?河北省衡水中學第一次綜合素養(yǎng)測評）Imaginethis.Youneedanimageofaballoonforawork

presentationandturntoanAItext-to-imagegeneratortocreateasuitableimage.Youentertheprompt（提示詞）“red

balloonagainstabluesky“butthegeneratorreturnsanimageofanegginstead.

Whafsgoingon?Thegeneratoryou'reusingmayhavebeen“poisoned”.Whatdoesthismean?Text-to-image

generatorsworkbybeingtrainedonlargedatasetsthatincludemillionsorbillionsofimages.Someofthegeneratorshave

beentrainedbyindiscriminatelyscrapingonlineimages,manyofwhichmaybeundercopyright.Thishasledtomany

copyrightinfringement（侵害）caseswhereartistshaveaccusedbigtechcompaniesofstealingandprofitingfromtheir

work.

Thisisalsowheretheideaof"poison“comesin.Researcherswhowanttoempowerindividualartistshaverecently

createdatoolnamed“Nightshade“tofightbackagainstunauthorisedimagescraping.Thetoolworksbyslightlyalteringan

image'spixels（像素）inawaythatconfusesthecomputervisionsystembutleavestheimageunalteredtoahuman?seyes.

IfanorganizationthenscrapesoneoftheseimagestotrainafutureAImodel,itsdatapoolbecomes"poisoned”.Thiscan

resultinmistakenlearning,whichmakesthegeneratorreturnunintendedresults.Asinourearlierexample,aballoonmight

becomeanegg.

Thehigherthenumberof“poisoned“imagesinthetrainingdata,thegreatertheimpact.Becauseofhowgenerative

AIworks,thedamagefrom^tpoisoned^^imagesalsoaffectsrelatedpromptkeywords.

Possibly,toolsHkeNightshadecanbeabusedbysomeuserstointentionallyupload"poisoned“imagesinorderto

confuseAIgenerators.ButtheNightshade'sdeveloperhopesthetoolwillmakebigtechcompaniesmorerespectfulof

copyright.Itdoeschallengeacommonbeliefamongcomputerscientiststhatdatafoundonlinecanbeusedforanypurpose

theyseefit.

Humanrightsactivists,forexample,havebeenconcernedforsometimeabouttheindiscriminateuseofmachine

visioninwidersociety.Thisconcernisparticularlyseriousconcerningfacialrecognition.Thereisaclearconnection

betweenfacialrecognitioncasesanddatapoisoning,asbothrelatetolargerquestionsaroundtechnologicalgovernance.It

maybebettertoseedatapoisoningasaninnovativesolutiontothedenialofsomefundamentalhumanrights.

12.Theunderlinedword“scraping”（para.2）isclosestinmeaningto.

A.facilitatingB.collectingC.damagingD.polishing

13.Accordingtothepassage,addingpoisoneddatamight.

A.increasetheaccuracyofreturnedinformation

B.leaduserstoforgetthepromptkeywords

C.causetroubletowiththetrainingofgenerativeAI

D.discriminateagainstgreatmasterpieces

14.Whatcanbeinferredfromthelasttwoparagraphs?

A.Datapoisoningissomehowjustifiedtodirectattentiontohumanrights.

B.Computerscientistshaslearnedtorespectthecopyrightofmostartists.

C.Nightshadeisbeingabusedbyhumanrightsactiviststorecognizefaces.

D.Theissueoftechnologicalgovernancehasarousedthelawyers5interest,

15.Whichofthefollowingmightbethebesttitleofthepassage?

A.DataPoisoning:GovernmentEmpoweringCitizenstoProtectThemselves

B.DataPoisoning:AddressingFacialRecognitionIssuesAmongArtists

C.DataPoisoning:RisksandRewardsofGenerativeAIDataTraining

D.DataPoisoning:RestrictingInnovationorEmpoweringArtists

【答案】12.B13.C14.A15.D

【解析】本文是一篇說明文。本文介紹了一種名為“Nightshade”的工具，該工具通過微妙地修改圖像的像素來混淆

計算機視覺系統(tǒng)，以此對抗未經授權的圖像抓取，保護藝術家的版權，同時引發(fā)了對數(shù)據中毒現(xiàn)象及其對人工智能

模型訓練影響的討論。

12.詞句猜測題。根據第二段劃線詞前面的句子“Text-to-imagegeneratorsworkbybeingtrainedonlargedatasetsthat

includemillionsorbillionsofimages.（文本到圖像生成器是通過在包含數(shù)百萬或數(shù)十億圖像的大型數(shù)據集上進行訓練

來工作的。）”可知，生成器是通過對大型數(shù)據集進行訓練來工作的，這些大型數(shù)據集包含著數(shù)百萬或者數(shù)十億圖像，

由此推斷生成器是通過收集這些數(shù)百萬或者數(shù)十億圖像進行訓練來工作的。同時根據第二段劃線詞后的“...online

images,manyofwhichmaybeundercopyright.Thishasledtomanycopyrightinfringement（侵害）caseswhereartists

haveaccusedbigtechcompaniesofstealingandprofitingfromtheirwork.（.......在線圖片，其中許多可能是受版權保護

的，這導致了許多侵犯版權的案件，藝術家們指責大型科技公司竊取他們的作品并從中獲利。）”可知，一些生成器

是通過不加選擇地抓取在線圖像來訓練的，其中許多圖像可能是受版權保護的，由此導致藝術家對大型科技公司進

行指責。由此可推理出劃線詞的含義為“收集”，故選B項。

13.細節(jié)理解題。根據第三段"Thetoolworksbyslightlyalteringanimage'spixels（像素）inawaythatconfusesthe

computervisionsystembutleavestheimageunalteredtoahuman,seyes.Ifanorganizationthenscrapesoneofthese

imagestotrainafutureAImodel,itsdatapoolbecomes“poisoned”.Thiscanresultinmistakenlearning,whichmakesthe

generatorreturnunintendedresults.（該工具通過以人類肉眼無法察覺、但能迷惑計算機視覺系統(tǒng)的方式，對圖像的像

素進行輕微修改。如果一個組織隨后抓取這些圖像來訓練未來的AI模型，其數(shù)據池就會變得“中毒”。這可能導致

錯誤的學習，使得生成器返回非預期的結果。）”可知，添加“中毒”數(shù)據可能干擾生成人工智能的訓練，故選C項。

14.推理判斷題。根據倒數(shù)第二段“Possibly,toolslikeNightshadecanbeabusedbysomeuserstointentionallyupload

ttpoisoned^^imagesinordertoconfuseAIgenerators.ButtheNightshade'sdeveloperhopesthetoolwillmakebigtech

companiesmorerespectfulofcopyright.（可能的是，像Nightshade這樣的工具可以被一些用戶濫用,故意上傳“有毒”

圖片以混淆AI生成器。但Nightshade的開發(fā)者希望這個工具能讓大型科技公司更加尊重版權。廠以及最后一段

“Humanrightsactivists,forexample,havebeenconcernedforsometimeabouttheindiscriminateuseofmachinevisionin

widersociety.（例如，人權活動家一直對機器視覺在更廣泛社會中的不加選擇使用表示關切）”可推理出，數(shù)據中毒在

某種程度上將人們的注意力引向人權是有道理的，故選A項。

15.主旨大意題。根據第——段“Youentertheprompt（提示詞）"redballoonagainstabluesky"butthegeneratorreturnsan

imageofanegginstead.（你輸入了提示詞“紅色氣球對比藍色天空”，但是生成器返回的卻是一張雞蛋的圖片）“可知，

文章提出了“生成器中毒的現(xiàn)象"；第二段“Thegeneratoryou'reusingmayhavebeen"poisoned”.（你正在使用的生成器

可能已經被"毒化"）"以及"Someofthegeneratorshavebeentrainedbyindiscriminatelyscrapingonlineimages,manyof

whichmaybeundercopyright.Thishasledtomanycopyrightinfringementcaseswhereartistshaveaccusedbigtech

companiesofstealingandprofitingfromtheirwork.”（一些生成器是通過不力口選擇地抓取在線圖像來訓|練的，其中許

多圖像可能屬于版權所有。這導致了許多侵犯版權的案件，藝術家指控大型科技公司竊取他們的作品并從中獲利）

不僅闡述了“生成器中毒”的問題，同時還提及了生成器是通過不加選擇地抓取在線圖像來訓練的，其中許多圖像引

發(fā)侵犯版權的問題；同時根據倒數(shù)第二段“Possibly,toolslikeNightshadecanbeabusedbysomeuserstointentionally

upload"poisoned“imagesinordertoconfuseAIgenerators.ButtheNightshade'sdeveloperhopesthetoolwillmakebig

techcompaniesmorerespectfulofcopyright.（可能的是，像Nightshade這樣的工具可以被一些用戶濫用，故意上傳“有

毒”圖片以混淆AI生成器。但Nightshade的開發(fā)者希望這個工具能讓大型科技公司更加尊重版權。）”以及最后一段

“Humanrightsactivists,forexample,havebeenconcernedforsometimeabouttheindiscriminateuseofmachinevisionin

widersociety.（例如，人權活動家一直對機器視覺在更廣泛社會中的不加選擇使用表示關切）”可知，此處強調

Nightshade的開發(fā)者呼吁該工具能讓大型科技公司更加尊重藝術創(chuàng)作者的版權。通讀全文，文章主要講述了人工智

能圖像生成器可能受到“毒害”的現(xiàn)象，即由于研究者為了保護藝術家版權而創(chuàng)造的工具Nightshade對圖像進行細微

像素修改，從而使得AI在訓練時產生錯誤學習，生成錯誤的圖像結果。這種“毒害”現(xiàn)象導致了許多侵犯藝術家版

權的案件，引起藝術家的不滿。由此文章提出質疑：到底該“中毒現(xiàn)象”是給藝術家賦權了還是限制了其創(chuàng)新能力。

D項“DataPoisoning:RestrictingInnovationorEmpoweringArtists（數(shù)據中毒：限制創(chuàng)新還是賦權藝術家）”表達的含義

適合用做本文標題，故選D項。

4.（2025?河南省洛陽市第一高級中學高三開學考試）Superhumanartificialintelligenceisalreadyamongus.Well,

sortof.WhenitcomestoplayinggameslikechessandGo,orsolvingdifficultscientificchallengeslikepredictingprotein

structures,computersarewellaheadofus.Butwehaveonesuperpowertheyaren'tclosetomastering:mindreading.

Humanshaveamysteriousabilitytoreasonthegoals,desiresandbeliefsofothers,avitalskillthatmeanswecan

predictotherpeople'sactionsandtheconsequencesofourown.Readingmindscomessoeasilytous,though,thatweoften

don'tthinktospelloutwhatwewant.IfAisaretobecometrulyusefulineverydaylife,wehavetogivethemthisgiftthat

evolutionhasgivenustoreadotherpeople'sminds.

Psychologistsrefertotheabilitytoinferother?smentalstateastheoryofmind.Inhumans,thisabilitystartsto

developataveryyoungage.Howtoreproducethecompetenceinmachinesisfarfromclear,though.Oneofthemain

challengesiscontext.Forinstance,ifsomeoneaskswhetheryouaregoingforarunandyoureply"it'sraining”,theycan

quicklyconcludethattheanswerisno.Butthisrequireshugeamountsofbackgroundknowledgeaboutrunning,weather

andhumanpreferences.Moreover,whetherhumansorAI,thetheoryofmindissupposedtoappearnaturallyfromone's

ownlearningprocess.

YetwemightstillwantAItohaveamorehuman-likeformoftheoryofmind.WhilelettingAIformthetheoryof

mindintheirlearningprocessislikelytoleadtodevelopingmorepowerfulAI,plainlybuildinginsharedwaystorepresent

knowledgemaybecrucialforhumanstotrustandcommunicatewithAL

Itisimportanttoremember,though,thatthepursuitofmachineswiththeoryofmindisaboutmorethanjustbuilding

moreusefulrobots.ItisalsoasteppingstoneonthepathtowardsadeepergoalforAIandroboticsresearch:buildingtruly

self-awaremachines.Whetherwewillevergetthereremainstobeseen,butweareonthepathtolearningtothinkabout

ourselves.

12.Whydoestheauthormention“games"and"proteinstructures^^inthefirstparagraph?

A.Tostresshowimportantthescienceis.B.TopromotetheuseofAIindailylife.

C.TopresentAFstheoryofmind.D.TocontrastwithAl'slackofmindreading.

13.Accordingtothepassage,whichofthefollowingcontextscanAIunderstandwell?

A.Whenkidsseetheirmomafterhurtingthemselvesandtheycrylouder.

B.Whenamomtellsherkidtoeatthefoodthatisgoodforhealthandheeatsit.

C.Whenateacherasksforaboy'shomeworkandheanswers“mydogateit”.

D.Whenyouareaskedtoeatspicyfoodfordinnerandyoureply"asorethroat".

14.Whatmaytheauthoragreewith?

A.Weshouldrejecthuman-likeformsofabilitiesforAL

B.Humans9theoryofmindisfarfromperfect.

C.Mindreadingrequireshugeamountsofcontext.

D.Thetheoryofmindisindependentofone'slearningprocess.

15.Whichwouldbethebesttitleforthepassage?

A.AIwithItsOwnTheoryofMindIsExpected

B.AIwithTheoryofMindWillReshapeOurFuture

C.AFsTheoryofMindIsaBlessingorSufferingtoHumans

D.TheoryofMindBridgestheGapBetweenHumansandAI

【答案】12.D13.B14.C15.A

【解析】本文是一篇議論文。文章討論了人工智能在模擬人類“讀心”能力方面的挑戰(zhàn)和前景。雖然AI在棋

類游戲和科學預測方面已超越人類，但缺乏理解他人目標和信念的能力。文章認為，為了讓AI在日常生

活中更有用，需要賦予其類似人類的“讀心”能力，即理論思維。這不僅是提高AI實用性的關鍵，也是實

現(xiàn)真正自我意識機器的重要步驟。

12.推理判斷題。根據第一段中“Superhumanartificialintelligenceisalreadyamongus.Well,sortof.Whenitcomesto

playinggameslikechessandGo,orsolvingdifficultscientificchallengeslikepredictingproteinstructures,computersare

wellaheadofus.Butwehaveonesuperpowertheyaren'tclosetomastering:mindreading.（超人的人工智能已經在我們

身邊了。嗯，算是吧。在下棋和圍棋等游戲，或解決預測蛋白質結構等困難的科學挑戰(zhàn)方面，計算機遠遠領先于我

們。但我們有一個超能力，他們還遠沒有掌握：讀心術）”可知，作者在第一段中提到人工智能擅長游戲和預測蛋白

質結構，也提到它還沒有掌握的的讀心術，將兩者進行對比。由此推知，作者在第一段中提到“游戲”和“蛋白質結

構”是為了與“人工智能缺乏讀心術”進行比較。故選Do

13.推理判斷題。根據第一段"Butwehaveonesuperpowertheyaren'tclosetomastering:mindreading.（但我們有一個

他們還沒有掌握的超能力：讀心術）”和第二段"Humanshaveamysteriousabilitytoreasonthegoals,desiresandbeliefs

ofothers,acrucialskillthatmeanswecananticipateotherpeople'sactionsandtheconsequencesofourown.（人類有一

種神秘的能力來推理他人的目標、欲望和信仰，這是一項至關重要的技能，意味著我們可以預測他人的行為和自己

的后果）”和第三段的“Forinstance,ifsomeoneaskswhetheryouaregoingforarunandyoureply"it'sraining^^,theycan

quicklyconcludethattheanswerisno.（例如，如果有人問你是否要跑步，而你回答“正在下雨”，他們很快就會得出結

論，答案是否定的）”可知，讀心需要經過一定的推理和預測能力，需要理論思維能力，而計算機不具備人類所擁

有的讀心的能力。對A項“Whenkidsseetheirmomafterhurtingthemselvesandtheycrylouder.（當孩子們在傷害自己

后看到他們的媽媽時，他們哭得更大聲）"、C項"Whenateacherasksforaboy'shomeworkandheanswers"mydogate

it”.（當老師問一個男孩的家庭作業(yè)時，他回答說“我的狗吃了和D項"Whenyouareaskedtoeatspicyfoodfbrdinner

andyoureply“asorethroat”.（你被要求晚餐吃辣的食物時，你回答說“喉嚨痛的理解都需要運用到思維理論去解釋

人們的目標和欲望,這是人工智能無法理解的。只有B項“Whenamomtellsherkidsomefoodisgoodforhealthandthe

kideatsit.（當媽媽告訴她的孩子一些食物對健康有益時，孩子就會吃）”是直接的陳述，不需要運用思維理論，人工

智能能很好的理解其意思。故選B。

14.推理判斷題。根據第三段中“Inhumans,thisabilitystartstodevelopataveryyoungage.Howtoreproducethe

competenceinmachinesisfarfromclear,though.Oneofthemainchallengesiscontext.Forinstance,ifsomeoneasks

whetheryouaregoingforarunandyoureply"it'sraining”,theycanquicklyconcludethattheanswerisno.Butthis

requireshugeamountsofbackgroundknowledgeaboutrunning,weatherandhumanpreferences.（在人類中，這種能力在

很小的時候就開始發(fā)展。然而，如何在機器中復制這種能力還遠不清楚。其中一個主要的挑戰(zhàn)是背景。例如，如果

有人問你是否要去跑步，你回答“下雨了”，他們可以很快得出結論，給出否定答案。但這需要大量關于跑步、天氣

和人類偏好的背景知識）”可知，理解別人的想法需要大量的背景信息。由此推知，作者會贊成“讀心術需要大量的背

景信息”這一觀點。故選C。

15.主旨大意題。通讀全文，結合最后一段的“Itisimportanttoremember,though,thatthepursuitofmachineswiththeory

ofmindisaboutmorethanjustbuildingmoreusefulrobots.Itisalsoasteppingstoneonthepathtowardsadeepergoalfor

AIandroboticsresearch:buildingtrulyself-awaremachines.（然而，重要的是要記住，追求具有心智理論的機器不僅僅

是為了制造更有用的機器人。這也是通往人工智能和機器人研究更深層目標的墊腳石：制造真正具有自我意識的機

器）”可知，文章討論了人工智能在模擬人類“讀心”能力方面的挑戰(zhàn)和前景，作者期待建造真正有自我意識的機

器，即作者期待人工智能具備思維理論。故A項“AIwithItsOwnTheoryofMindIsExpected（期待人工智有擁有自己

的心智理論）”可以作為本文的最佳標題。故選A。

5.（2025?湖北省騰云聯(lián)盟高三上學期8月聯(lián)考）Researcherssaythey'veusedcutting-edgegravitationalwave

researchtocastnewlightonamystery-a2000-year-oldcomputer,theAntikytheramechanismfoundinshipwreck.

Welloveracenturyafteritsdiscovery,researchersattheUniversityofGlasgowsaythey'veusedstatisticalmodeling

techniques,originallydesignedtoanalyzegravitationalwaves-ripplesinspacetimecausedbymajoreventsinthe

universesuchastwoblackholescombining-tosuggestthattheAntikytheramechanismwaslikelyusedtotracktheGreek

lunaryear.Inshort,ifsafascinatingcollisionbetweenmodern-dayscienceandthemysteriesofanancientartifact.

Ina2021paper,researchersfoundthatpreviouslydiscoveredandregularlyspacedholesina"calendarring^^were

markedtodescribethe“motionsofthesun,moon,andallfiveplanetsknowninancientGreeksandhowtheywere

displayedatthefrontasanancientGreekuniverse.^^Now,inanewstudypublishedintheOfficialJournaloftheBritish

HorologicalInstitute,UniversityofGlasgowgravitationalwaveresearcherGrahamWoanandresearchassociateJoseph

Bayleysuggestthattheringwaslikelyperforated（打孔）with354holes,whichhappenstobethenumberofdaysinalunar

year.

Theteamusedstatisticalmodelsderivedfromgravitationalwaveresearch,alarge-scalephysicsexperimentdesigned

tomeasureripplesinspacetimemillionsoflight-yearsfromEarthandBayesiananalysis,atechniqueusingprobabilityto

quantifyuncertaintybasedonincompletedata,tocalculatethelikelynumberofholesinthemechanismusingthepositions

ofthesurvivingholesandtheplacementofthering'ssurvivingsixfragments.

Surprisingly,theinspirationforthestudycamefromaYouTuberChrisBudiselic,whohasbeenattemptingto

physicallyrecreatetheancientmechanismandinvestigatingwaystodeterminejusthowmanyholesitcontained.

"It'saneatsymmetrythatwe'veadaptedtechniquesweusetostudytheuniversetodaytounderstandmoreabouta

mechanismthathelpedpeoplekeeptrackoftheheavensnearlytwomillenniaago,“Woansaid.

“WehopethatourfindingsabouttheAntikytheramechanism,althoughlesssupematurallyspectacularthanthose

madebyI