Previously ATLAS-R-002
DatasetPreprintOpen Access
KiranaChain: A Multi-Agent Negotiation Benchmark for India's Informal Retail Economy
- Dhadi Sai Praneeth Reddy1
- 1 Atlas AI Labs
01
Brief
- Problem
- Agent negotiation benchmarks rarely reflect informal economies, where credit is extended on trust, supply is irregular and trade happens across languages.
- Approach
- KiranaChain models India's kirana-store economy as a multi-agent negotiation environment covering informal credit, supply-chain dynamics and multilingual commerce.
- Contribution
- An openly released benchmark on Hugging Face (DOI 10.57967/hf/9015) for evaluating negotiating agents in a setting that matters to 12–14 million stores.
02
Status
Preprint
- Idea
- Experiments
- Drafting
- Review
- Released (current stage)
03
Abstract
Agent negotiation benchmarks rarely reflect informal economies, where credit is extended on trust, supply is irregular and trade happens across languages. KiranaChain models India's kirana-store economy as a multi-agent negotiation environment covering informal credit, supply-chain dynamics and multilingual commerce. An openly released benchmark on Hugging Face (DOI 10.57967/hf/9015) for evaluating negotiating agents in a setting that matters to 12–14 million stores.
04
Identifiers
- DOI
- 10.57967/hf/9015Copy DOI
05
Key results
12–14M
Kirana stores modelled
India's informal retail economy.
3
Dynamics modelled
Informal credit · supply chains · multilingual commerce.
06
Write-up
Why kirana stores
India's 12–14 million kirana stores run on relationships: credit extended informally, stock that arrives irregularly, and conversations that move between languages. These are exactly the conditions under which negotiating agents should be tested.
What the benchmark models
- Informal credit between stores and customers
- Supply-chain dynamics between stores and suppliers
- Multilingual commerce across the languages trade actually happens in
07
Cite this work
08
