REVIEW 9 cited by
Interactive AI Alignment: Specification, Process, and Evaluation Alignment
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Interactive AI Alignment: Specification, Process, and Evaluation Alignment
read the original abstract
Modern AI enables a high-level, declarative form of interaction: Users describe the intended outcome they wish an AI to produce, but do not actually create the outcome themselves. In contrast, in traditional user interfaces, users invoke specific operations to create the desired outcome. This paper revisits the basic input-output interaction cycle in light of this declarative style of interaction, and connects concepts in AI alignment to define three objectives for interactive alignment of AI: specification alignment (aligning on what to do), process alignment (aligning on how to do it), and evaluation alignment (assisting users in verifying and understanding what was produced). Using existing systems as examples, we show how these user-centered views of AI alignment can be used descriptively, prescriptively, and as an evaluative aid.
Forward citations
Cited by 9 Pith papers
-
User identity conditions moral wrongness ratings in non-reasoning large language models
Implicitly conveying a user's professional role in multi-turn LLM conversations shifts moral wrongness ratings across ten common-morality rules in two non-reasoning models.
-
Intentmaking and Sensemaking: Human Interaction with AI-Guided Mathematical Discovery
Expert mathematicians using an AI coding agent for discovery engage in repeated cycles of intentmaking to define goals and sensemaking to interpret outputs.
-
Deployment-Relevant Alignment Cannot Be Inferred from Model-Level Evaluation Alone
Deployment-relevant AI alignment cannot be inferred from model-level evaluations alone, as benchmark audits show missing interaction support and cross-model tests reveal model-dependent scaffold effects.
-
MultEval: Supporting Collaborative Alignment for LLM-as-a-Judge Evaluation Criteria
MultEval supports collaborative creation of LLM-as-a-judge criteria by surfacing disagreements via consensus-building methods, allowing iterative revisions with examples and history, and keeping transparent how human ...
-
Co-Constructing Alignment: A Participatory Approach to Situate AI Values
Misalignments appear in practice as unexpected responses and task breakdowns, with users proposing roles such as adjusting model output, interpreting behavior, or deliberate non-use to co-construct alignment.
-
LLMs Get Lost In Multi-Turn Conversation
LLMs drop 39% in performance during multi-turn conversations due to premature assumptions and inability to recover from early errors.
-
How Creatives Approach GenAI Image Generation: Tensions Between Structured Guidance, Self-Experimentation, and Creative Autonomy
Creatives prefer self-experimentation over structured guidance for GenAI image tools to preserve creative freedom, even when guidance aids AI literacy.
-
Value-Sensitive AI for Prayer: Balancing the Agencies Between Human and AI Agents in Spiritual Context
AI systems risk reducing the perceived authenticity of prayer when they take too much guiding agency, so designs should preserve user agency through interpretive openness or allow non-use.
-
How Creatives Approach GenAI Image Generation: Tensions Between Structured Guidance, Self-Experimentation, and Creative Autonomy
Qualitative studies show creatives prefer self-experimentation over structured guidance for GenAI image tools to preserve creative autonomy despite terminology barriers.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.