List your company

Instruction Following on ALFWorld (test)

91.4Success RateMemP
-1.6822.48546.6570.815Jun 29, 2026Jun 30, 2026Jul 1, 2026Jul 2, 2026Jul 3, 2026

Evaluation results

MethodLinks
MemPModel=Qwen2.5-72B (Teacher)2026.0691.4——
No Mem.Model=Qwen2.5-72B (Teacher)2026.0687.1——
DuoMemModel=Qwen3-4B-Inst. (4B)2026.0677.973.622.9
+LoRAModel=Qwen3-4B-Inst. (4B)2026.0672.1——
DuoMemModel=Gemma3-12B-it (12B)2026.0666.452.125
MemPModel=Qwen3-8B (8B)2026.0664.3——
DuoMemModel=Qwen3-8B (8B)2026.0664.323.60
+CDModel=Qwen3-8B (8B)2026.0661.4——
+LoRAModel=Qwen3-8B (8B)2026.0660.7——
+CDModel=Qwen3-4B-Inst. (4B)2026.0656.4——
+LoRAModel=Gemma3-12B-it (12B)2026.0656.4——
DuoMemModel=Gemma4-E2B-it (2B)2026.0655.753.641.4
aTTTSignal=Self, Model=Qwen3.5-9B2026.0755.7——
aTTTSignal=Env, Model=Qwen3.5-9B2026.0755.5——
MemPModel=Qwen3-4B-Inst. (4B)2026.0655——
aTTTSignal=Summary, Model=Qwen3.5-9B2026.0754.3——
Sequence filterSignal=Summary, Model=Qwen3.5-9B2026.0754——
Sequence filterSignal=Self, Model=Qwen3.5-9B2026.0753.8——
No filterSignal=Self, Model=Qwen3.5-9B2026.0753.1——
qTTTModel=Qwen3.5-9B2026.0752.4——
Sequence filterSignal=Env, Model=Qwen3.5-9B2026.0752.1——
No filterSignal=Summary, Model=Qwen3.5-9B2026.0752.1——
No filterSignal=Env, Model=Qwen3.5-9B2026.0751——
ReActModel=Qwen3.5-9B2026.0750.7——
+LoRAModel=Gemma4-E2B-it (2B)2026.0646.4——
+CDModel=Gemma3-12B-it (12B)2026.0646.4——
MemPModel=Gemma3-12B-it (12B)2026.0641.4——
No Mem.Model=Qwen3-8B (8B)2026.0640.7——
aTTTSignal=Summary, Model=Gemma-3-12B2026.0726.7——
Sequence filterSignal=Summary, Model=Gemma-3-12B2026.0726.2——
No filterSignal=Summary, Model=Gemma-3-12B2026.0725.5——
aTTTSignal=Env, Model=Gemma-3-12B2026.0725——
aTTTSignal=Self, Model=Gemma-3-12B2026.0724.3——
Sequence filterSignal=Self, Model=Gemma-3-12B2026.0723.8——
Sequence filterSignal=Env, Model=Gemma-3-12B2026.0722.9——
qTTTModel=Gemma-3-12B2026.0722.6——
ReActModel=Gemma-3-12B2026.0722.4——
No filterSignal=Self, Model=Gemma-3-12B2026.0722.1——
No filterSignal=Env, Model=Gemma-3-12B2026.0721.7——
+CDModel=Gemma4-E2B-it (2B)2026.0616.4——
MemPModel=Gemma4-E2B-it (2B)2026.0614.3——
No Mem.Model=Gemma3-12B-it (12B)2026.0614.3——
Sequence filterSignal=Env, Model=Qwen3.5-4B2026.075.5——
aTTTSignal=Summary, Model=Qwen3.5-4B2026.074.8——
aTTTSignal=Env, Model=Qwen3.5-4B2026.074.5——
No filterSignal=Summary, Model=Qwen3.5-4B2026.074.5——
No Mem.Model=Qwen3-4B-Inst. (4B)2026.064.3——
No filterSignal=Env, Model=Qwen3.5-4B2026.074.3——
Sequence filterSignal=Summary, Model=Qwen3.5-4B2026.074.3——
aTTTSignal=Self, Model=Qwen3.5-4B2026.073.6——
Sequence filterSignal=Self, Model=Qwen3.5-4B2026.073.1——
qTTTModel=Qwen3.5-4B2026.072.6——
No Mem.Model=Gemma4-E2B-it (2B)2026.062.1——
No filterSignal=Self, Model=Qwen3.5-4B2026.072.1——
ReActModel=Qwen3.5-4B2026.071.9——