PoliCon: LLM Evaluation on Political Consensus Tasks

PoliCon is a benchmark accepted at ICLR 2026 for evaluating large language models on political consensus tasks. It simulates a European Parliament setting where models must reconcile multi-party political stances on policy topics. The dataset was created by Yofuria and last updated in March 2026.

Use Cases

• Benchmarking LLM performance on political negotiation tasks based on simulated multi-party stances.
• Training models for consensus-building in parliamentary settings based on the described European Parliament simulation.
• Analyzing model alignment with diverse political objectives and voting rules as described in the benchmark.

Strengths

• The dataset is associated with a benchmark accepted at a major conference (ICLR 2026).
• It provides a structured simulation environment based on a European Parliament setting.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Yofuria

Freshness

Last updated 2026-03-06 06:36:44; freshness should be verified.

Geography

Likely focuses on European political contexts.