List your company

Multi-hop Question Answering on 2WikiMultiHopQA (Token-level F1)

65.9Token F1 ScorePDA-GAM
49.46853.7345862.266May 29, 2026

Evaluation results

MethodLinks
PDA-GAMStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=42026.0565.9
PDA-GAMStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=32026.0565.6
PDA-TIESStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=42026.0565.4
PDA-TIESStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=32026.0565.2
PDA-AveragingStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=42026.0564.9
PDA-KnOTSStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=42026.0564.7
PDA-GAMStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=42026.0564.2
PDA-TIESStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=42026.0564
PDA-AveragingStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=42026.0563.8
PDA-TIESStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=32026.0563.8
PDA-GAMStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=32026.0563.8
PDA-KnOTSStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=42026.0563.6
PDA-GAMStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=22026.0563.5
PDA-TIESStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=22026.0563.4
PDA-TIESStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=22026.0562.8
Search-R1Student Model=Qwen3-8B, Preference Delta Scale=7B over 3B, Number of Preference Deltas (∆)=12026.0562.6
PDA-GAMStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=22026.0562.4
Joint ORPOStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=42026.0562
Qwen3Student Model=Qwen3-8B, Preference Delta Scale=4B over 1.7B, Number of Preference Deltas (∆)=12026.0561.5
Search-R1Student Model=Tülu3-8B, Preference Delta Scale=7B over 3B, Number of Preference Deltas (∆)=12026.0561.3
Joint ORPOStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=42026.0560.6
AceSearcherStudent Model=Qwen3-8B, Preference Delta Scale=7B over 3B, Number of Preference Deltas (∆)=12026.0560.1
Qwen3Student Model=Tülu3-8B, Preference Delta Scale=4B over 1.7B, Number of Preference Deltas (∆)=12026.0559.2
Llama-3.2Student Model=Qwen3-8B, Preference Delta Scale=3B over 1B, Number of Preference Deltas (∆)=12026.0558.5
Student BaselineStudent Model=Qwen3-8B2026.0558.1
AceSearcherStudent Model=Tülu3-8B, Preference Delta Scale=7B over 3B, Number of Preference Deltas (∆)=12026.0557.4
Llama-3.2Student Model=Tülu3-8B, Preference Delta Scale=3B over 1B, Number of Preference Deltas (∆)=12026.0555.8
Student BaselineStudent Model=Tülu3-8B2026.0555.1
Sequential ORPOStudent Model=Qwen3-8B, Number of Preference Deltas (∆)=22026.0552.2
Sequential ORPOStudent Model=Tülu3-8B, Number of Preference Deltas (∆)=22026.0550.1