teleo/teleo-codex
8
0
Fork 0
Code Issues 22 Pull requests 14 Projects Releases Packages Wiki Activity Actions 65

theseus: extract from 2025-11-00-operationalizing-pluralistic-values-llm-alignment.md

Browse source 
- Source: inbox/archive/2025-11-00-operationalizing-pluralistic-values-llm-alignment.md
- Domain: ai-alignment
- Extracted by: headless extraction cron (worker 5)

Pentagon-Agent: Theseus <HEADLESS>
This commit is contained in:
Teleo Agents 2026-03-12 08:54:53 +00:00
parent ba4ac4a73e
commit a3404b7229
1 changed files with 14 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
inbox/archive/2025-11-00-operationalizing-pluralistic-values-llm-alignment.md
View file

@ -7,9 +7,14 @@ date: 2025-11-01
domain: ai-alignment domain: ai-alignment
secondary_domains: [] secondary_domains: []
format: paper format: paper
status: unprocessed status: null-result
priority: high priority: high
tags: [pluralistic-alignment, demographic-composition, empirical, safety-inclusivity, real-human-feedback] tags: [pluralistic-alignment, demographic-composition, empirical, safety-inclusivity, real-human-feedback]
processed_by: theseus
processed_date: 2026-03-11
enrichments_applied: ["community-centred norm elicitation surfaces alignment targets materially different from developer-specified rules.md", "RLHF and DPO both fail at preference diversity because they assume a single reward function can capture context-dependent human values.md", "some disagreements are permanently irreducible because they stem from genuine value differences not information gaps and systems must map rather than eliminate them.md"]
extraction_model: "anthropic/claude-sonnet-4.5"
extraction_notes: "First large-scale empirical study quantifying the effect of demographic composition on alignment outcomes. Single high-confidence claim extracted with three enrichments to existing pluralistic alignment claims. Effect sizes (3-5 pp) are substantial and cannot be dismissed as noise with N=1,095."
--- ---
## Content ## Content
@ -37,3 +42,11 @@ Demonstrates that "whose feedback" matters as much as "how much feedback" for al
PRIMARY CONNECTION: community-centred norm elicitation surfaces alignment targets materially different from developer-specified rules PRIMARY CONNECTION: community-centred norm elicitation surfaces alignment targets materially different from developer-specified rules
WHY ARCHIVED: Empirical evidence that "whose preferences" is a quantitatively important question, not just a fairness concern WHY ARCHIVED: Empirical evidence that "whose preferences" is a quantitatively important question, not just a fairness concern
EXTRACTION HINT: Focus on the magnitude of demographic composition effects and what this means for single-population alignment training EXTRACTION HINT: Focus on the magnitude of demographic composition effects and what this means for single-population alignment training
## Key Facts
- Study included 27,375 ratings from 1,095 participants
- Liberal feedback improved models 5.0 percentage points vs Conservative baseline
- White feedback improved models 4.7 percentage points vs Black baseline
- Female feedback improved models 3.4 percentage points vs Male baseline
- Effects measured on emotional awareness and toxicity dimensions