diff --git a/paper.aux b/paper.aux index 1e9de08..a766460 100644 --- a/paper.aux +++ b/paper.aux @@ -3,14 +3,14 @@ \providecommand\HyField@AuxAddToFields[1]{} \providecommand\HyField@AuxAddToCoFields[2]{} \@writefile{toc}{\contentsline {section}{\numberline {1}Conditioning, Not Memory}{1}{section.1}\protected@file@percent } -\@writefile{toc}{\contentsline {section}{\numberline {2}Method}{1}{section.2}\protected@file@percent } -\@writefile{toc}{\contentsline {subsection}{\numberline {2.1}Architecture}{1}{subsection.2.1}\protected@file@percent } +\@writefile{toc}{\contentsline {section}{\numberline {2}Method}{2}{section.2}\protected@file@percent } +\@writefile{toc}{\contentsline {subsection}{\numberline {2.1}Architecture}{2}{subsection.2.1}\protected@file@percent } \@writefile{toc}{\contentsline {subsection}{\numberline {2.2}Conditioning}{2}{subsection.2.2}\protected@file@percent } \@writefile{toc}{\contentsline {subsection}{\numberline {2.3}Triggering}{2}{subsection.2.3}\protected@file@percent } \@writefile{toc}{\contentsline {subsection}{\numberline {2.4}Why Hidden States, Not Text}{2}{subsection.2.4}\protected@file@percent } -\@writefile{toc}{\contentsline {section}{\numberline {3}Experiments}{2}{section.3}\protected@file@percent } -\newlabel{sec:results}{{3}{2}{Experiments}{section.3}{}} -\@writefile{toc}{\contentsline {subsection}{\numberline {3.1}Setup}{2}{subsection.3.1}\protected@file@percent } +\@writefile{toc}{\contentsline {section}{\numberline {3}Experiments}{3}{section.3}\protected@file@percent } +\newlabel{sec:results}{{3}{3}{Experiments}{section.3}{}} +\@writefile{toc}{\contentsline {subsection}{\numberline {3.1}Setup}{3}{subsection.3.1}\protected@file@percent } \@writefile{toc}{\contentsline {subsection}{\numberline {3.2}One-Shot Conditioning}{3}{subsection.3.2}\protected@file@percent } \@writefile{lot}{\contentsline {table}{\numberline {1}{\ignorespaces Post-bias degeneration correlates with both instruct tuning and architectural complexity (sliding window attention, KV sharing, logit softcapping). Confounded in current test matrix.}}{3}{table.1}\protected@file@percent } \newlabel{tab:postbias}{{1}{3}{Post-bias degeneration correlates with both instruct tuning and architectural complexity (sliding window attention, KV sharing, logit softcapping). Confounded in current test matrix}{table.1}{}} @@ -18,16 +18,17 @@ \@writefile{lot}{\contentsline {table}{\numberline {2}{\ignorespaces Cross-model discrimination. Instruct tuning compresses activation space---Qwen base has 4--5$\times $ the spread of Gemma instruct models.}}{3}{table.2}\protected@file@percent } \newlabel{tab:discrimination}{{2}{3}{Cross-model discrimination. Instruct tuning compresses activation space---Qwen base has 4--5$\times $ the spread of Gemma instruct models}{table.2}{}} \@writefile{toc}{\contentsline {subsection}{\numberline {3.4}Quantization Tolerance}{3}{subsection.3.4}\protected@file@percent } -\@writefile{lot}{\contentsline {table}{\numberline {3}{\ignorespaces Actual quantized inference (bitsandbytes, Qwen). Same-precision self-match is always 1.000. Cross-precision f32$\to $int4 drops to 0.944.}}{3}{table.3}\protected@file@percent } -\newlabel{tab:quant}{{3}{3}{Actual quantized inference (bitsandbytes, Qwen). Same-precision self-match is always 1.000. Cross-precision f32$\to $int4 drops to 0.944}{table.3}{}} \citation{jang2024camelot} \citation{fountas2024emllm} \citation{das2024larimar} -\citation{lewis2020rag} -\citation{meng2022rome,meng2023memit} +\@writefile{lot}{\contentsline {table}{\numberline {3}{\ignorespaces Actual quantized inference (bitsandbytes, Qwen). Same-precision self-match is always 1.000. Cross-precision f32$\to $int4 drops to 0.944.}}{4}{table.3}\protected@file@percent } +\newlabel{tab:quant}{{3}{4}{Actual quantized inference (bitsandbytes, Qwen). Same-precision self-match is always 1.000. Cross-precision f32$\to $int4 drops to 0.944}{table.3}{}} \@writefile{toc}{\contentsline {section}{\numberline {4}Privacy by Representation}{4}{section.4}\protected@file@percent } \@writefile{toc}{\contentsline {section}{\numberline {5}Related Work}{4}{section.5}\protected@file@percent } -\@writefile{toc}{\contentsline {section}{\numberline {6}Limitations}{4}{section.6}\protected@file@percent } +\@writefile{toc}{\contentsline {subsection}{\numberline {5.1}Behavioral Conditioning}{4}{subsection.5.1}\protected@file@percent } +\@writefile{toc}{\contentsline {subsection}{\numberline {5.2}Training-Free External Memory}{4}{subsection.5.2}\protected@file@percent } +\citation{lewis2020rag} +\citation{meng2022rome,meng2023memit} \bibstyle{plainnat} \bibcite{das2024larimar}{{1}{2024}{{Das et~al.}}{{}}} \bibcite{fountas2024emllm}{{2}{2024}{{Fountas et~al.}}{{}}} @@ -36,9 +37,11 @@ \bibcite{meng2022rome}{{5}{2022}{{Meng et~al.}}{{}}} \bibcite{meng2023memit}{{6}{2023}{{Meng et~al.}}{{}}} \bibcite{pavlov1927}{{7}{1927}{{Pavlov}}{{}}} +\@writefile{toc}{\contentsline {subsection}{\numberline {5.3}Other Approaches}{5}{subsection.5.3}\protected@file@percent } +\@writefile{toc}{\contentsline {section}{\numberline {6}Limitations}{5}{section.6}\protected@file@percent } +\@writefile{toc}{\contentsline {section}{\numberline {7}Conclusion}{5}{section.7}\protected@file@percent } \bibcite{raz2005}{{8}{2005}{{Raz et~al.}}{{}}} \bibcite{skinner1938}{{9}{1938}{{Skinner}}{{}}} \bibcite{weitzenhoffer1957}{{10}{1957}{{Weitzenhoffer}}{{}}} \bibcite{hypnosis1930}{{11}{1930}{{Hypnosis \& CR}}{{}}} -\@writefile{toc}{\contentsline {section}{\numberline {7}Conclusion}{5}{section.7}\protected@file@percent } -\gdef \@abspage@last{5} +\gdef \@abspage@last{6} diff --git a/paper.log b/paper.log index 1bdbb91..a918430 100644 --- a/paper.log +++ b/paper.log @@ -1,4 +1,4 @@ -This is pdfTeX, Version 3.141592653-2.6-1.40.29 (TeX Live 2026/Arch Linux) (preloaded format=pdflatex 2026.3.11) 6 APR 2026 19:50 +This is pdfTeX, Version 3.141592653-2.6-1.40.29 (TeX Live 2026/Arch Linux) (preloaded format=pdflatex 2026.3.11) 6 APR 2026 19:51 entering extended mode restricted \write18 enabled. %&-line parsing enabled. @@ -428,29 +428,33 @@ LaTeX Font Info: Trying to load font information for U+msb on input line 23. File: umsb.fd 2013/01/14 v3.01 AMS symbols B ) [1 -{/var/lib/texmf/fonts/map/pdftex/updmap/pdftex.map}] -Overfull \hbox (41.58173pt too wide) in paragraph at lines 120--124 +{/var/lib/texmf/fonts/map/pdftex/updmap/pdftex.map}{/usr/share/texmf-dist/fonts +/enc/dvips/cm-super/cm-super-ts1.enc}] +Overfull \hbox (41.58173pt too wide) in paragraph at lines 124--128 \OT1/cmr/m/n/10.95 it (1536-dim, 35 lay-ers), E4B base (2560-dim, 42 lay-ers). Quan-ti-za-tion tested at f32/f16/bf16/int8/int4 [] -[2{/usr/share/texmf-dist/fonts/enc/dvips/cm-super/cm-super-ts1.enc}] [3] -[4] [5] (./paper.aux) +[2] + +LaTeX Warning: `h' float specifier changed to `ht'. + +[3] [4] [5] [6] (./paper.aux) *********** LaTeX2e <2025-11-01> L3 programming layer <2026-01-19> *********** Package rerunfilecheck Info: File `paper.out' has not changed. -(rerunfilecheck) Checksum: 4198C661EF4EF302CF6CA8FE934684BD;2020. +(rerunfilecheck) Checksum: 8536E375995754CA41E5F9B69D548498;2546. ) Here is how much of TeX's memory you used: - 12083 strings out of 467525 - 178010 string characters out of 5425861 - 607900 words of memory out of 5000000 - 40829 multiletter control sequences out of 15000+600000 + 12090 strings out of 467525 + 178106 string characters out of 5425861 + 607903 words of memory out of 5000000 + 40832 multiletter control sequences out of 15000+600000 639085 words of font info for 84 fonts, out of 8000000 for 9000 1141 hyphenation exceptions out of 8191 - 75i,8n,79p,324b,520s stack positions out of 10000i,1000n,20000p,200000b,200000s + 75i,8n,79p,324b,567s stack positions out of 10000i,1000n,20000p,200000b,200000s -Output written on paper.pdf (5 pages, 196804 bytes). +Output written on paper.pdf (6 pages, 198036 bytes). PDF statistics: - 210 PDF objects out of 1000 (max. 8388607) - 171 compressed objects within 2 object streams - 42 named destinations out of 1000 (max. 500000) - 121 words of extra memory for PDF output out of 10000 (max. 10000000) + 227 PDF objects out of 1000 (max. 8388607) + 187 compressed objects within 2 object streams + 46 named destinations out of 1000 (max. 500000) + 145 words of extra memory for PDF output out of 10000 (max. 10000000) diff --git a/paper.pdf b/paper.pdf index bb9202a..5ee9886 100644 Binary files a/paper.pdf and b/paper.pdf differ diff --git a/paper.tex b/paper.tex index 275ae22..c910971 100644 --- a/paper.tex +++ b/paper.tex @@ -48,9 +48,13 @@ the model is exactly as it was---no trace, no residue. The closer analogy is post-hypnotic suggestion: a trigger installed externally, fired without the subject's awareness, removable without leaving a mark. -Fine-tuning modifies weights. RAG re-encodes text each time. LoRA requires -gradients. In-context learning vanishes with the conversation. CRI persists -across sessions without touching the model. +\begin{itemize} +\item Fine-tuning modifies weights and causes catastrophic forgetting. +\item RAG re-encodes text each time --- no persistent behavioral change. +\item LoRA requires gradients. +\item In-context learning vanishes with the conversation. +\item CRI persists across sessions without touching the model. +\end{itemize} %─────────────────────────────────────────────── \section{Method} @@ -130,7 +134,7 @@ diverges: \begin{table}[h] \centering -\begin{tabular}{llll} +\begin{tabular}{lll} \toprule Backbone & Post-bias behavior & Fluent? \\ \midrule @@ -212,27 +216,44 @@ operating in the model's internal space. %─────────────────────────────────────────────── \section{Related Work} -\textbf{Pavlov}~(1927) described hypnotic suggestion as the best example of a -conditioned reflex in humans. \textbf{``Hypnosis and the Conditioned Reflex''} -(1930) formalized this: suggestion installs stimulus-response links that fire -without the subject's awareness. CRI implements the same mechanism on -transformers: activation pattern (CS) paired with logit biases (US) produces -token sequence (CR). \textbf{Raz et~al.}~(2005) showed post-hypnotic suggestion -reduces conflict in human brains by modulating activity in specific -regions---external behavioral modification without awareness, analogous to -CRI's logit injection. \textbf{Skinner}~(1938): operant conditioning. CRI -currently performs respondent conditioning only; bias modulation via reward is a -natural extension. +\subsection{Behavioral Conditioning} -\textbf{CAMELoT} \citep{jang2024camelot}: KV pairs from attention, injected as -prefixes. \textbf{EM-LLM} \citep{fountas2024emllm}: KV cache extension. -\textbf{Larimar} \citep{das2024larimar}: memory matrix, requires training. -All inject at attention level. CRI injects at output logits---simpler, cheaper, -no attention recomputation. +\begin{itemize} +\item \textbf{Pavlov}~(1927) described hypnotic suggestion as the best example + of a conditioned reflex in humans. +\item \textbf{``Hypnosis and the Conditioned Reflex''}~(1930) formalized this: + suggestion installs stimulus-response links that fire without awareness. +\item \textbf{Raz et~al.}~(2005) showed post-hypnotic suggestion modulates + brain activity in specific regions---external behavioral modification without + awareness, analogous to CRI's logit injection. +\item \textbf{Skinner}~(1938): operant conditioning. CRI currently performs + respondent conditioning only; bias modulation via reward is a natural + extension. +\end{itemize} -\textbf{RAG} \citep{lewis2020rag}: retrieves text, re-encodes. RAG informs; CRI -conditions. \textbf{ROME/MEMIT} \citep{meng2022rome,meng2023memit}: rank-one -weight edits. CRI modifies zero weights. +CRI implements the Pavlovian mechanism on transformers: activation pattern (CS) +paired with logit biases (US) produces token sequence (CR). + +\subsection{Training-Free External Memory} + +\begin{itemize} +\item \textbf{CAMELoT} \citep{jang2024camelot}: KV pairs from attention, + injected as prefixes. +\item \textbf{EM-LLM} \citep{fountas2024emllm}: KV cache extension. +\item \textbf{Larimar} \citep{das2024larimar}: memory matrix, requires training. +\end{itemize} + +All inject at the attention level. CRI injects at output logits---simpler, +cheaper, no attention recomputation. + +\subsection{Other Approaches} + +\begin{itemize} +\item \textbf{RAG} \citep{lewis2020rag}: retrieves text, re-encodes into + context. RAG informs; CRI conditions. +\item \textbf{ROME/MEMIT} \citep{meng2022rome,meng2023memit}: rank-one weight + edits. CRI modifies zero weights. +\end{itemize} %─────────────────────────────────────────────── \section{Limitations}