NishiSearch

Web This site Trusted

30 result(s) for “okapi bm25 site hotchips pp site github com site re if site www ncid nl nil site:arxiv.org” of 39 matched · 631 ms

Scope: Nishi’s own index (our crawl + Common Crawl) — not the whole live web. Some sites block all automated crawlers (anti-bot walls), so their pages may be absent here even though we never filter lawful content.

Did you mean okapi bm25 site hotchips pp site github com site re if site www nid nl nil site:arxiv.org?

Freshness: no result on this page carries a fetch time yet (indexed before the fetch epoch was stored beside the doc)

The query asks the model to recall the topic of an earlier conversation about Tesla; the gold answer is anchored in one specific session about Pooja’s dig site
from the top result below

Report GitHub Issue

https://arxiv.org/html/2608.02613v1
We evaluate five open-weight readers with Vanilla context, BM25-RAG, Oracle retrieval, Memobase, and MemSearch as memory backends

^ |ɒKl  v?q|rr0

https://arxiv.org/src/2006.01354
~D(;32-fDj#cB@@nl[IU,Ud0>\_`TBx 1-sBi|Ƙ4̛Ɯ4ȫkTv;ǓJBDe zfLf#2:HZQ`:1=%TM9{P&#G%H=):ICa NPg

rj;srY9靟OO߰ޟ4'}Ʊڅ'o1d{l

https://arxiv.org/src/2608.03883
ޛco 6t`Zk9BAJus7jTD#'96e7S˜=)*3uD ҁ j2ϯIVn7t=M];L6t R(ɤtŤeI>D0]Zp]S wm_} ONMKvs:0aw"&T Θo|4{Cb1/Yl}װ6BMI R7эP/W_,nL0hYؙ ·p"P˟=R5E(A*=x,{QV'Q

%PDF-1.5 % 181 0 obj > stream xZKs8Wp dfbOav*[p!

https://arxiv.org/pdf/2005.14165
%bbK9JVx]}gQHՠKdy˪nL%",DF4f QR,=O澞6}s=iDo

%PDF-1.5 % 152 0 obj > stream xڭ[Yܸ~ׯЃ:Fh8^X?

https://arxiv.org/pdf/2102.13654
48#vBB҈ mt={v ֳĉ{Wkv_rORu>YS\+V,LqO8--gϞMqJLLTwv͛7O߷o_pp8Mҥ >uּSN +V,]kVre ; y)񄿀pW[Y'0A=zxX,={VfE$0Q[JH*4Z|do߾

w_z߯x;ã'x|p ãGsJĊ(2w[

https://arxiv.org/src/2006.02230
2+45(@;KQ&5HS+DD~:Q7X0]""n@h|P_a&%Bqr^݊'|V}W4 nGo>h(F?)SxCƗz^G_A +|CHurX$=ҜnR ߵ-TSZ^a } /_SS`]2f|m0Ui@S75Ј4

%PDF-1.5 % 152 0 obj > stream xmTˎ0+$$0  a#A%

https://arxiv.org/pdf/2402.03300
߯jD岻fc;Z̫MfG} q]/ޭmޯo⣩0Z^x]fkn{E+{*ʧypg6;5PVpH8$hmڢ*߄zR:")󨺠3qXysO'H)-"}[˺s 3 4{pYdrK+ a }ѫW{ Fvm7344AGc

m_\ysFϿO[d턄RdTSxW3RMIAfϾqrnI

https://arxiv.org/src/2006.02464
>>!ŧ/h4>!BO28w/ #,l4I~˥ ]Jo ?t&;Iԋiz3f6-O4}z,(L$MD4~\4J7K%(@WubL,4| bG bU\j)cL W^ SKB;

8kd[[S ՙ9 ;n_2e !!/n٭

https://arxiv.org/src/2211.01910v2
-wBS3Cb---뷖Dr5GO ?]󉟵^}v[~j9N},?ipOZ[HLdL|1@3ZV Cy')տ8[F:0/10IQi|}/OnP0 E EGZi ž0>ދ 6Ў[h}RXc25

J^P_.I/"]Mzi"H =[@*H.Z!t{

https://arxiv.org/src/2006.01402
O~ЋqzQnSbZoCu$ U#/y`htNL7/|Zr'ϧkȝ

%PDF-1.7 % 1 0 obj > stream xVM6W.`[CQIZT

https://arxiv.org/pdf/2608.03233
ܖue51(,aaIx +%^Lpxm“&4KCV7ֽy qCj":A|YpG;hSH_;238xSxstNlNNB{2w0w_w|^ Al0Δ2 l|d~GPޮA}qYNKޙu5\RTaX+g'haid(G7y5#d'Bp֪'~UP_0V_3 "pݐ~eME-m*0V>8

rj[rFϿ#!Y|X=J9W+Cr,@0I\S$8

https://arxiv.org/src/2608.03882
'|"ѪbH~>ȈRE⍊$*q䵼b9,*y[p#q-wܦ 7$aG+k)3ǹvzL ѰCE)FI2]jN2L Srkەb(z{gEZDj s*ݫw:RvZmۚvɮS:"@Æh}|c|'k12|=m{W,a羽 "W8q ]᠛$d0 $6͈/,á~Ku]>k+/aytDhDJqUԙV :

%PDF-1.5 % 1 0 obj > stream xY]o6}_fuo ( ,omE}ZADږ

https://arxiv.org/pdf/2608.02851v1
%6G"8/1tx493C`LԒ)c9S18YecܨՌ;4LCς eQ0F3pLVVһL -ѳc8zt4h(SdSLՂ5\ҳaZ*LkC|r*1)3I83Ҥ) hL1F2aYebt,qU1+(nřG.|-;tϾvcynЏ

%PDF-1.5 % 1 0 obj > endobj 2 0 obj > endobj 3 0 obj > stream

https://arxiv.org/pdf/2508.16279
AgentScope 1.0: A Developer-Centric Framework for Building Agentic ApplicationsDawei GaoZitao LiYuexiang XieWeirui KuangLiuyi YaoBingchen QianZhijian MaYue CuiHaohao LuoShen LiLu YiYi YuShiqi HeZhiling LuoWenmeng ZhouZhicheng ZhangXuguang

Report GitHub Issue

https://arxiv.org/html/2608.03048v1
com/JetAstra/PI-Mem Introduction The ability to reason over long contexts is increasingly important for large language models (LLMs), supporting applications such as long-document question answering (Gemini Team 2025; Yang et al

1 Introduction

https://arxiv.org/html/2501.13956v1
com Abstract We introduce Zep, a novel memory layer service for AI agents that outperforms the current state-of-the-art system, MemGPT, in the Deep Memory Retrieval (DMR) benchmark

1 Introduction

https://arxiv.org/html/2407.16741v3
com/All-Hands-AI/OpenHands Slack http://bit

Artificial Intelligence

https://arxiv.org/list/cs.AI/new
Biological systems, in contrast, exhibit adaptive plasticity, maintaining function through dynamic re-organisation around damage

%PDF-1.7 % 1 0 obj > stream xYێ6}߯cC; gf憽HKߪ

https://arxiv.org/pdf/2608.00785
*"4X]滉` Yd)$w(P\w󲒴e$S IGRǼ%:VͶg{*Ɗъ$vbyxZ1phN"l̀ >=9͟^11>"mS;wn *j[Gx.ڛ7pi

%PDF-1.7 % 1 0 obj > stream xYYܸ~ϯ`fěv,l^Bfw

https://arxiv.org/pdf/2608.01785
B%pp7w}^t3>/nN y@IvonG

z;c:XSW._U20eKrI$!B\UkQPk+V

https://arxiv.org/src/2205.11916v3
>FLE#񫲉E#| @8W 0X52Ž7lDYGݷ&O8àz,51INQfm]TNl(fb$fd\ >$vULw‹7̰%1

%PDF-1.5 % 2 0 obj > stream xmTˎ0+$$0  a#A%

https://arxiv.org/pdf/2407.02402
TD_97;'2>I|zP"~o7A5lM$du\%z{LUiWǕQre3VZȨm˶z+MuO%ߞ[[}nJwչRh " |"N1-ܿ|uþ04)`YHcD)BQEN}O=q"ƒf:JRjq>

%PDF-1.5 % 153 0 obj > stream xڭ]o=B,\>q$6@DD

https://arxiv.org/pdf/2109.12196
$!8C{gvfdݥ/9ooz!/NEB^aBſZ ])A仲-Q  ȧ(޽~qzw旲g׹?^}H/0KdEk)T27ąL?̴@Oe@W޲xe{ZԂ76r͢6MPMP4, #г}3Bn^

^sx ]7 Ƕm۶m۶mۘLlۙ؞X_~~\

https://arxiv.org/src/2006.00441
m7Tg`[nL~pQ]O#T) yqRH0k10

%PDF-1.5 % 1 0 obj > endobj 2 0 obj > endobj 3 0 obj > stream

https://arxiv.org/pdf/2506.01844
SmolVLA: A Vision-Language-Action Model for Affordable and Efficient RoboticsMustafa ShukorDana AubakirovaFrancesco CapuanoPepijn KooijmansSteven PalmaAdil ZouitineMichel AractingiCaroline PascalMartino RussiAndres MarafiotiSimon

%PDF-1.5 % 162 0 obj > stream x[KƑϯ`"vD7:#Y5hD

https://arxiv.org/pdf/2407.00132
Pvͬ*Tഴ@*_f%& Qnʤd܈&K2%6Ϳm3͝4I]=Cj>Uz=қL|KH7gII7{&MTYlmFjކfwqueŶT{zƕd:ZcdzVE3`]Ano3@r)+/Ԭn`kYfy

%PDF-1.7 % 1 0 obj > stream x[is۸~?Nޔ bRuV%v8N

https://arxiv.org/pdf/2608.02615
mNJ3mDbwF=Ml;x,qq,ޣ^}ZXy:wWW|

Computation and Language

https://arxiv.org/list/cs.CL/new
We evaluate five open-weight readers with Vanilla context, BM25-RAG, Oracle retrieval, Memobase, and MemSearch as memory backends

[2608.03883] DS@GT-ARC at eRisk 2026 Task 3: Sparse, Semantic, and LLM

https://arxiv.org/abs/2608.03883
Our systems combine sparse BM25 retrieval, evidence-aware rescoring for self-referential symptom reports, embedding-based reranking, query-prototype expansion, and LLM-based reranking

[2104.08663] BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of

https://arxiv.org/abs/2104.08663
Our results show BM25 is a robust baseline and re-ranking and late-interaction-based models on average achieve the best zero-shot performances, however, at high computational costs