Leaderboard

Final test-phase standings. Each team is ranked by its best valid submission, and every score was independently re-checked with the official scorers before release.

#	Team	Accuracy ↑	Balanced Acc	Macro-F1
1	am2401295	0.9620	0.9332	0.9081
2	NYUAD	0.9610	0.9193	0.8718
3	CUET_InferX	0.9120	0.8772	0.8050

Open on Codabench

#	Team	Accuracy ↑	Balanced Acc	Macro-F1
1	NYUAD	0.8750	0.7799	0.7265
2	am2401295	0.8340	0.7211	0.6856
3	Digilians	0.6560	0.5745	0.5044

Open on Codabench

#	Team	Contrastive Instability ↓	Combined Acc	Q+ Acc	Q− Acc
1	miles_ma	0.0290	0.9710	0.9710	0.9855
2	myamin66	0.0330	0.9670	0.9670	0.9835
3	mohaiminulhoque	0.0350	0.9650	0.9650	0.9825
4	NYUAD	0.0410	0.9590	0.9590	0.9795
5	hasankh	0.0510	0.9490	0.9490	0.9745
6	sumaiya_110	0.0580	0.9420	0.9420	0.9710
7	md_faisal	0.0670	0.9330	0.9330	0.9665
8	keslerady	0.1130	0.8870	0.8870	0.9435

filtered: keslerady · run 876194 · no overlapping ids

Open on Codabench

#	Team	Contrastive Instability ↓	Combined Acc	Q+ Acc	Q− Acc
1	ahmedbakr	0.0360	0.9640	0.9640	0.9820
2	NYUAD	0.0470	0.9530	0.9530	0.9765
3	myamin66	0.0560	0.9440	0.9440	0.9720
4	hasankh	0.0960	0.9040	0.9040	0.9520
5	md_faisal	0.1010	0.8990	0.8990	0.9495
6	keslerady	0.1420	0.8580	0.8580	0.9290

filtered: ahmedbakr · run 883723 · no overlapping ids

Open on Codabench

#	Team	Spearman (composite) ↑	MAE (composite)
1	md_faisal	0.8258	0.1469
2	nannies	0.8078	0.1512
3	2231171736	0.8040	0.1415
4	ahmedbakr	0.7806	0.1452
5	am2401295	0.6734	0.2032

Open on Codabench

Test phase closed 10 August 2026, 12:00 UTC (August 9, AoE).

Standings from the development phase, scored on the devtest split. These were interim results; the test phase is the official ranking.

#	Team	Accuracy ↑	Balanced Acc	Macro-F1
1	NYUAD	0.9740	0.9743	0.9740
2	NYUAD	0.9740	0.9743	0.9740
3	romey101	0.9720	0.9718	0.9717
4	CUET_InferX	0.9160	0.9160	0.9157
–	hunzed baseline	0.6640	0.6658	0.6614

Open on Codabench

#	Team	Accuracy ↑	Balanced Acc	Macro-F1
1	NYUAD	0.9200	0.9188	0.9211
2	NYUAD	0.9200	0.9188	0.9211
3	Digilians	0.7840	0.7799	0.7823
–	hunzed baseline	0.3980	0.4101	0.3444
4	md_faisal	0.3980	0.4101	0.3444

Open on Codabench

#	Team	Contrastive Instability ↓	Combined Acc	CFHR	Q+ Acc	Q− Acc
1	mohaiminulhoque	0.0280	0.9720	0.0000	0.9720	0.9860
2	NYUAD	0.0280	0.9720	0.0000	0.9720	0.9860
3	miles_ma	0.0300	0.9700	0.0000	0.9700	0.9850
4	nawwad	0.0360	0.9640	0.0000	0.9640	0.9820
5	Nile Nexus	0.0500	0.9500	0.0000	0.9500	0.9750
6	ameya	0.2840	0.7160	0.0000	0.7160	0.8580
–	hunzed baseline	0.3133	0.6840	0.2723	0.9400	0.8390
7	md_faisal	0.3133	0.6840	0.2723	0.9400	0.8390

filtered: NAMAA Community · run 875076 · no overlapping ids

Open on Codabench

#	Team	Contrastive Instability ↓	Combined Acc	CFHR	Q+ Acc	Q− Acc
1	NYUAD	0.0360	0.9640	0.0000	0.9640	0.9820
2	NAMAA Community	0.0460	0.9540	0.0000	0.9540	0.9770
3	lettycat	0.0820	0.9180	0.0000	0.9180	0.9590
–	hunzed baseline	0.4900	0.5080	0.3909	0.8340	0.7730
4	md_faisal	0.4900	0.5080	0.3909	0.8340	0.7730

Open on Codabench

#	Team	Spearman (composite) ↑	MAE (composite)
1	ahmedbakr	0.7076	0.1876
2	md_faisal	0.6413	0.2036

Open on Codabench

Final development-phase snapshot, taken 3 August 2026, 10:55 UTC.