Rather than telling a dog what I don't want it to do, (which is certainly appropriate at other times) when I'm trying to train a specific behavior I'd rather show my dog what I DO want him to do. Leash corrections aren't really doing that for you.
Reinforcing the dog for walking calmly on a loose leash next to me works much better, IMO.
I start training my pups to walk next to me around the house, off leash at first. Hand targeting is great for this, and I'll also lure with a treat. A hallway works well because you can use the wall as a guide to proper positioning. I also use the dog run, which is a narrow concrete side yard with a wood fence along one side and the house on the opposite side. Not much in the way of distractions, so it's a good next step.
I start out very slowly, working up to several steps in a row. Three steps is basically the beginning, middle and end of the behavior of loose leash walking, but one method I used with Halo was referred to as the wedding march - take one step, stop. Take another step, stop. Rinse, repeat. When I stop, I want the dog to automatically sit at my side, so I'll lure that with a treat, then mark (with a clicker or a verbal marker - "yes!"), and give the treat. My loose leash walking cue is "let's go". When we're out on a walk I do like to give my dogs permission to check stuff out, so I put that on cue. If they start to pull towards something they want to smell, I make them sit and give me eye contact, then I release to "go sniff". I take them for long leashed hikes (3-10 miles at a time), so I don't require a perfect heel with the dog's head aligned with my let, but I do have specific criteria that I train for and expect.
I used several methods on this link: How to Train a Dog to Stop Pulling on the Leash | Grisha Stewart
I really like the "silky leash" method, which used to be described in detail on Grisha's website along with videos, but now that she has a book out there's not much information there anymore. What it does is train the dog to yield to light pressure of the leash rather than pulling against it, and you begin training it in a very low distraction environment. Here's a video I found, his timing is a bit slow, but you can see the general idea:
I also used #2 from her site - reward for eye contact, #3 - reward for being in the sweet spot, and #4 - the canine cha-cha, which works well for penalty yards. The more the dog pulls towards something, the further away he gets. Walk nicely, we go forward, pull and we go backwards. You can practice this around the house with a bowl and some dog treats as bait, or a favored toy. When you get close to the bowl or toy and he will sit and look at you, you can pick up whatever the thing is and give it to him. This is also an exercise in impulse control. Obviously, you would do this on leash, to prevent him from getting at the bait.
I also like and have used Helix Fairweather's shaping method. There are 3 steps. I found the first step too rudimentary after the work I'd already put in with Halo, but I think it's worth looking at if you've never seen it before - it even features a GSD! I think it works particularly well if the dog already understands the concept of leash pressure, which Halo did due to silky leash, but the beginning of the first video shows the dog excited and pulling at first:
I skipped right to this step:
As you'll see, after taking a step backwards and pivoting, the dog is at her side in heel position. I actually jumped ahead at this point and rather than stopping after the pivot I took several steps with Halo at my side.
This is what it looks like after the foundation work is done. The rate of reinforcement is still high, and you can see how engaged the dog is with her. One note about not being interested in treats outdoors, they may not be high value enough, and you may be starting in too distracting of an environment. What are you using for treats?